~BNFL Buried Concrete at WETS-A Literature Review 1-p -
BNFL Inc.
A Literature Review on The Behaviour of .Buried Contaminated Concrete Over Time and Relevance to Plutonium, Americium and Uranium Contamination at The Rocky Flats Environmental Technology Site.
Submitted to Rocky Mountain Remediation Services
PO Box 464 Rocky Flats Environmental technology Site CO 80402-0464
BY BNFL Inc
6143 South Willow Drive, Suite 430 Englewood CO 801 11 @BNR Buried Concrete at RFETS-A Litentun Review I"*
Contents
EXECUTIVE SUMMARY 4
INTRODUCTION 5
Overview of radionucliddcement literature 6
Literature search criteria 7
NATURE OF CONTAMINATION OF ROCKY FLATS CONCRETE 8
SITE SPECIFIC GEOCHEMISTRY 8
Summary of Rocky Flats Geochemistry 8
Variations in Specific Parameters 9 Carbonate 9 Sulphate 9 Chloride 10 Reduced Nitrogen, Sulphur and Organic Matter 10 10 PH Redox 10
CEMENT AND CONCRETE DEGRADATION 11
introduction 11
Chemical evolution of cement minerals 11 Hydrated cement minerals 12 Leaching of Ca(OH), and CSH 12
Carbonation 14
Sulphate, Magnesium and Chloride attack 15
Alkali - aggregate reactions 16
Corrosion of steel reinforcements 17
Biodeterioration of Concrete 17 Microbially Induced Degradation of Concrete and Cement 18 Generation of Complexing Agents 20 Factors lnfluencing Biodeterioration 20 Water availability 20 PH 20 Nutrients 21 Temperature 21 Presence of Micro-organisms 21
Page 2 of 57 @BNRInc . Burred Concrete at WETS-A Literature Review
Colloid Generation in Cementitious Systems 21
Ancient analogues of concrete degradation 22
Repository degradation models 22
Concrete degradation relevant to radionuclide mobility at Rocky Flats 23
LEACHING OF ACTINIDES IN CEMENTITIOUS SYSTEMS 25
Conclusions on Leaching Experiments 26
BEHAVIOUR OF ACTINIDES IN CEMENTITIOUS ENVIRONMENTS 27
Summary of Conditions expected in a cementitious environment 27
Speciation of Actinides 28 General 28 Plutonium 29 Uranium 31 Americium 31
Solubility of actinides in cementitious environments 31 Plutonium 31 Uranium 34 Americium 37
Sorption of actinides onto cements 37
SUMMARY AND CONCLUSIONS 41
Concrete Degradation 41
Leaching, solubility and sorption of plutonium, uranium and americium in cementitious systems 42
BIBLIOGRAPHY 45
Page 3 of 57 @BNFLInc Buned Concrete rt WETS-A Literatwe Revlew
Executive Summary
Thls report reviews current literature on the behaviour of plutonium, uranium and amencium in contaminated concrete wth relevance to surface contarmnated concrete at the Rocky Flats Environmental Technology Site (WETS) This review wll be used in the second part of the project to design a drafi work plan to more fully understand the mobility of thls contammahon at WETS for the purpose of assessing radiologic nsk over a penod of 1000 years The aim of the overall project is to provide a companson of the behaviour of plutonium, uranium and americium present in contaminated concrete to that present in soils for whch nsk cntena are defined The project w11 also establish what concentration of these radionuclides can be bound in concrete at WETS that w11 be protective of water quality standards
A review of literature on cement and concrete degradation, radionuclide leaching from cement and concrete, and plutonium, uranium and amencium solubility and sorption under cementitious conditions has been undertaken Sufficient information is avsulable in the literature and in the WETS Sitewde Geoscience Charactenzation study to qualitatively predict the degradation behaviour of concrete rubble at WETS and to provide background information for evaluating the chemical controls on leaching, solubility and sorption of plutomum, uranium and amencium The man process of degradation affecting the predominant surf. contamination of WETS concrete is likely to be carbonation, the reaction wth carbon dioxide either in soil gases or dissolved in groundwater Attack by sulphate and processes controlled by microbial activity may also be relevant, but the involvement of these processes is more uncertam at present Corrosion of steel rebar is also likely to contribute to the cracking of concrete blocks The rnsun form of degradation by carbonation is unlikely to result in dissaggregation of the surface and may in fact result in a more resistant surface which is less prone to fieeze-thaw and mechanical erosion Formation of silica and iron hydroxide colloids is possible dmng degradation which could influence the migration of actinides
Carbonation of the surface zone contamng plutonium, urmum and amencium contamination has unportant implications for the chemical controlled leaching of radionuclides The local fluid pH surroundmg the contamination w11 be lowered fiom above 12 to - 8 by carbonation Redox (Eh) condihon wll be unaffected by carbonation and may not be changed from that of the background geochemistry, steel corrosion is the only likely mechanism by which reducing conditions may be established It is emphasised that these conditions are not typical of the extreme high pH, low Eh deliberately engineered in radioactive waste repository designs and thus plutonium, uranium and amencium are likely to be more mobile in WETS concrete than in such repositories
Expenmental studies of leaching of plutonium, uranium and americium from cementitious materials are very limited This is largely a result of the low mobility of actinides under these conditions, and the need for long time scale experiments Data that is available would confirm the proposition that liquid contamination as well as particulate contamination is very close to the surface Under these circumstances difisive control is less important as the contamination wll be in close contact with groundwater Leaching from surface contamination is considered to be controlled by processes of sorption onto the degraded cement surface and, where concentrations are sufficiently high, by solubility control from a solid contaminant (e g PuO,) In view of thn, a review of literature on plutomum, urafllum and amencium solubility and sorption under conditions in the degraded concrete surface has been undertaken A substantial amount of literature is avsulable concemng the solubility of actmdes under cementitious conditions, although the majonty of these studies are concerned Page 4 of 51 @BNFL . - Buried Concrete at WETS-A Literature Revim Inc
An possible important factor controllmg solubility, and hence release is the nature and solubility of the actinide contamination Sorption distribution coefficients (Rds or Kds) for actimdes under lugh pH conditions are qwte well defined, but again these manly relate to hgh pH, non-carbonated cement substrates Actinide sorption is generally less effectwe at lower pH, and uncertanties exist in the sorption mechanism on carbonated cements and on the uptake ofcontaminants by calcium carbonate
In summary, a reasonable understanding ofthe degradation of concrete at WETS is possible, the question of the mobility of plutomum, uranium and amencium under the chemical conditions established in the degraded concrete is more uncertain These contaminants are likely to be solubility controlled in the very near surface ofthe degraded concrete, however the nature and solubility limit ofthe solid contaminant phases are unknown
Introduction
This report compnses a review of the literature on processes relevant to the mobility of the radionuclides plutonium, uranium and americium present in contaminated concrete present at the Rocky Flats Environmental Technology site (RFETS) It has been suggested, for the purposes of assessing radiological nsk, that release of radionuclides from contaminated concrete rubble remaimng after decommissiomng actiwties at WETS w11 be similar to that from release from soil over a 1,000 year penod (RMRS, 1998) On this basis a possible radiation dose based standard for determining the concentration of contaminants in concrete is the Tier 1 soil action level defined in the Rocky Flats Cleanup Agreement (RFCA) (DOE, 1998) Another possible standard is that radionuclides leached from concrete wll be protective of water quality defined by Tier 1 action levels Am for ground water, Pu, 15 pCiL ,U 100 pCiL (DOE, 1998, Attachment 5)
The aims of this project as defined in the statement of work (RMRS, 1998) is to provide a qualitative assessment of the physical and chemical properties of plutonium, amencium and urmum contamination in concrete and to answer the questions “Does concrete behave like soils at WETS over the 1,000 year assessment penod7”. “What concentration of plutomum, amencium and uranium may be bound in concrete left at RFETS after decommissioning that wll be protective of water quality standards7”
This literature review is the first stage ofthis exercise, the objectives ofthe review are to provide a preliminary understanding of the processes involved in concrete degradation and plutonium, americium and urmum leaching from concrete, at WETS This review includes examination of site specific data provided pnncipally in the Groundwater Geochemistry Report for the RFETS (EG&G, 1995), together with verbal descnptions of the nature of the contamination in concrete from RMRS and Kaiser-Hi11 staff A qualitative description of the degradation of concrete at RFETS and the chemical factors governing the mobility of plutonium, amencium and uranium are provided Comments are made on current inadequacies in the open literature in understanding processes ofactinide mobility from contaminated concrete at RFETS
In the second stage of the project gaps in the literature wll be defined and a draft work plan will be designed to fill missing data and to test the environmental effects ofconcrete behaviour in the environment In the draft work plan report, a conceptual model ofthe release ofplutonium, amencium and uranium from concrete left at Page 5 of 51 @BNFLInc Buried Concnte at WETS-A Literature Review WETS w11 be produced Avalable literature data wll be used to assess the concentration of these radionuclides wthin contaminated concrete that are protective of water quality
Overview of radionucliddcement literature
An extensive amount of literature exists in peer-reviewed jo~rnals,conference proceedings and in reports of vmous agencies and research institutes on the application of cement and concrete to the disposal of radioactive waste Cement-based matenals are used commonly for the encapsulation of low- and intermediate-level waste and for proposed backfill material in repository designs Concrete is used extensively in the structural design of radioactive waste repositories Cement and concrete form a low permeability bamer to the ingress of groundwater to radioactive waste In addition, the high pH and hgh surface area of the cement matnx provides a chemical bamer to radionuclides, providing retardation by sorption and lowenng solubility To understand the behaviour of hschemical barrier extensive research has been camed out to predict the evolution of chemical conditions in cement pore water over very long penods of time (100,000 years) Much of hsresearch has concentrated on predicting pH dmng leachmg of cement by groundwater as this is an important parameter controlling both the solubility and sorption of radionuclides Other areas of cement degradation, which impact on the structural integnty of concrete are attack by chlonde, sulphate and magnesium, which can be at high concentrations in deep subsurface bnnes The microbial promoted attack of concrete is also quite frequently discussed in the radioactive waste literature
Measurement of leaching rate of certain radionuclides from cement wasteforms has been carried out Quite extensive expenmental studies have been made of Cs, Co, Sr, I, C leaching (Amoya and Suzuki, 1992, Ktrshnamoorthy et al, 1992, Plecas et al, 1992b, Miyamoto et al, 1993, Nishi et al, 1991, Kat0 et al, 1996, Penc et al, 1993, 1994, 1995) and models based on diffusive transport of these relatively mobile elements have been developed (Plecas et a1 1992a, Kim et al, 1992, 1993, 1996) Such data and models are not particularly relevant to the leaching of contaminants fiom WETS concrete, here the contamination is in the form of Pu, Am and U, and the surficial nature of the contamination wll not be influenced by diffusive properties of the cement matnx
Studies of the leachmg of actinides from cementitious matenals are much more limited, hsm part reflects the very low mobility of actinides in cementitious systems, wth the consequence that expenmental time scales have to be extremely long In addition, the focus of this limited work has beer, concerned with the leachmg of radionuclides from cement-encapsulated waste, rather than surface contamination Diffusion measurements do however provide a useful indicator of the depth to which actinides can penetrate into concrete
Despite the lack of data explicitly dealing wth actinide leaching from concrete, there is another avenue to be explored Due to the use and proposed use of cementitious matenal for the immobilisation of radioactive waste, properties of radionuclides in cement environments have been extensively studied Thus solubilities of actinides in cement leachates have been determined, and sorption onto cement pastes have been measured Although these do not, on the face of it, appear to be directly relevant to WETS leach behaviour, these experiments may in fact reveal useful information about the leaching of radionuclides from cements
As a consequence of the literature search cntena used, and the number of references related to radioactive waste disposal, many references examined relate to the use of cement based matenals in repository designs and waste encapsulation It should perhaps be emphasised at th~sstage that although the literature reviewed is Page 6 of 57 @BNFL . - Buncd Concrctc at WETS-A Literature Revlew Inc sourced largely fiom such repository studies, the processes controlling plutomum, amencium and uranium mobility at WETS are not typical of those in radioactive waste repositones In particular
e Contamination in WETS concrete is sdicial and w11 be in close contact wth groundwater pH wll be significantly lower than in engineered repositones, where it forms an integral part of the multi- bmer approach
L terature search criteria
The literature search was cmed out using BNFL’s Information Retneval Service The followng databases were searched
BIOSIS Previews CA SEARCH (Chemical Abstracts) CAE3 ABSTRACTS Ceramic Abstracts Chemical Engineenng and Biotechnology Abstracts Dement World Patents Index DIALOG Sourceone Engineenng Dissertation Abstracts Online E1 Compendex Plus Electnc Power Database EMBASE Energy, Science and Technology (DOE) Envnoline FLUIDEX GeoArchive GEOBASE GeoRef Inside conferences INSPEC JICST - Eplus Life Sciences Collection METADEX NTIS Nuclear Science Abstracts PASCAL SciSearch SPIN TOXLINE TRIS Wilson Applied Science World Translations Index
The followng keyword searches were made 1 Cement or concrete and leach and radionuclide (1 92 returns) 2 Cement or concrete and degradation (1 56 returns) 3 Concrete and contamination and radionuclides or uranium or plutonium or fission product or nuclear waste or radioactive waste (95 returns)
These searches were selected so that the key publications on radionuclide mobility in cementitious radioactive waste systems would be covered, the relevance of these publications was determined as site specific data was received from RMRS The last search concerning concrete contamination was the least successful and references here mainly concerned decontamination methods, contamination associated wth nuclear reactors and nuclear fall out On the basis of document title and avilable abstracts 60 papers and documents were chosen as being relevant and were obtained from the British Lending Library In addition to these references a number of other sources of literature were examined which included
Page 7 of 57 OBNFL . - Buried Concrete at WETS-A Literature Rev~m Inc
1 Matenals Research Society Symposia “ Scientific Basis for Nuclear Waste Management” VIII-XXI (1 985- 1998) 2 A collection ofreferences on actimde solubility and sorphon
In addition reference lists in reviewed publications were exammed for further relevant references A list of all references used in th~sreview is provided in the Bibliography
Nature of Contamination of Rocky Flats Concrete
Information regarding the nature of contamination of construction concrete at WETS is limited and no documented descnptions were avalable Verbal descnptions of the nature of the contamination were provided at two meetings (Roberts, personal commwcation, Ervin, personal commumcation) Contamination of Am WETS construction concrete by Pu, U and is thought to be present in three forms
0 Surface contamination ofvery fine grained PuO, from a plutonium fire 0 Contamination from Pu mtrate solution seepage through concrete 0 Uranium present in locally denved aggregate used in concrete
The surface contaminated Pu is thought to be the most important for discussion here since contamination by Pu nitrate may be at levels too high for in-situ disposal Plutonium is likely to be present in the chemically stable form PuO,, as very fine-grained smoke particles resulting from combustion of Pu metal (Carnal1 and Choppin, 1983) Surface deposited PuO, particulates in concrete are generally likely to be limited to the first few millimetres (DOE, 1995) Urmum and amencium are assumed to be present on the surface in a similar form to surficial Pu, although the nature of these compounds is more uncertain than for plutonium, uraruum could be present as either UO, or in the U(V1) oxidahon state as hydrated phases (e g schoepite) if contamination is related to combustion, amencium may be present as hydroxide The depth of liqud mtrate contamination on WETS concrete is not known Expenments have however been performed on the Am attenuahon of Pu and mobility in concrete immersed in mtrate (Jakubick,’ 1987) whch show that at a Am Am steady state uptake from 10” M Pu and 10“ M solutions both Pu and penetrated to a depth of around 2cm in concrete pre-treated wth 3M nitnc acid Pu diffusion and leachmg expenments reviewed in a later section indicate that diffusion is extremely slow Liquid Pu mtrate contamination is therefore likely to be close to the surface, and thus wll be influenced by chemical degradation of the concrete Uranium in aggregate is present presumably at background levels in local bedrock Uranium in aggregate is assumed to be evenly distributed through the concrete and its release wll be controlled by the gross degradation of the concrete
Site Specific Geochemistry
The geochemistry and composition of the groundwater at WETS is a crucial factor in determimng the leaching behaviour of surface contaminated concrete Therefore, a review ofthe groundwater geochemistry is appropnate This review is based on the comprehensive site charactensation report (EG+G, 1995), and it has not be deemed necessary to offer any new interpretation Rather, those factors of groundwater geochemistry Page 8 of 57 @BNFL Burred Concrete at WETS-A Lttcratun Review Inc that are important, from a concrete leachmg context, have been highlighted, extracted and assessed The am is to summanse the magnitude and variahon of important parameters, in order for the literature review to be placed wthin a site specific context
Summary of Rocky Flats Geochemistry
The hydrology of the WETS site is divided into two distinct utllts, the Upper Hydrostratigraphic Unit (UHSU) and the Lower Hydrostratigraphc Umt (LHSU) The UHSU is composed of unconsolidated, surficial deposits and weathered sandstone of the underlying Laramie and Arapahoe formations The LHSU consists of the unweathered portions of the Arapahoe and Laramie formations
Groundwater flow is generally from west to east, wth permeability and hydraulic conductivities greater in the UHSU than in the LHSU There appears to be limited hydrological commutllcation between the two units The vertical hydraulic conductwity of the LHSU is an order of magnitude less than the honzontal conductivity, indicating that the LHSU acts as a hydraulic bmer to vertical groundwater flow The groundwater geochemistry report (EG&G, 1995) presents data from samples collected from 1990 and 1994, from 532 wells The data is presented in terms of Stiff and Piper diagrams, and time senes graphs In addition, the results from geochemical modelling (using WATEQF) and inverse modellingheaction path modelling (using NETPATH) are also shown and discussed
In summary, the UHSU and LHSU compnse significant geochemical units, wth the former predominantly calcium bicarbonate dominated and the latter considerably more vaned, ranging from a sodium bicarbonate to sodium sulphate dominated groundwater A major exception to these generalities is found in groundwaters close to the Operational Units, where contamination results in UHSU compositions as varied as LHSU waters, wth sodium and sulphate particularly enhanced
Data from wells along four proposed flow paths are also presented, and also interpreted through modelling In general, the concentrations of major cations (Ca, Mg, Na and IC) and total dissolved solids increase along the flow paths, whch could be indicative of contamination, infiltration or mixing of groundwaters However, inverse modelling suggests that the evolution of groundwater composition can be explaned by natural geochemical reactions, such as dissolution and precipitation of mineral phases, and ion exchange reactions In particular, the precipitation of kaolinite, SiO, and iron minerals, the dissolution of calcite, pyrite, microcline and chlonte and exchange of calcium for sodium, would appear to be the more important mechanisms The conclusion is that the mixing of additional groundwaters is not required to account for the evolution of major element geochemistry along the major flow paths
However, local vanations are shown to exist, particular along the Industrial Area flow path, where the major ion composition of the UHSU changes from a calcium bicarbonate water to a mixed sodium bicarbonate/sodium sulphate groundwater In particular, groundwaters close to the Solar Evaporation Ponds indicate elevated levels of major cations, strontium, uranium-235, chlonde and sulphate, while groundwaters close to the Landfill exhibit low pH's, potentially low Eh and elevated metal concentrations It does not appear that these local vanations are reflected in the overall geochemical evolution along the flow paths, as the groundwater at the end ofthe flow path is again a calcium bicarbonate water
Page 9 of 57 @BNFL Bumd Concrete at WETS-A Literature RCVICW Inc -
Variations in Specific Parameters
Carbonate
Carbonation is an important degradation mechanism, which affects the physical and chemical integnty of concrete Carbonate also affects the solubility and sorption of actimdes in groundwater As has been mentioned above, the groundwaters of the UHSU can be classified as calcium bicarbonate, wth the carbonate level likely to be controlled by calcite solubility Takmg data from the Flow Path diagrams (Figures 5-8 to 5- 17, EG&G, 1995), bicarbonate concentrations appear to vary from less than 1 mgA in the background wells of the Southern Flow Path to 200 mg/l along the Woman Creek flow path However, boreholes in the Industnal area exhlbit concentrabons much higher than hs For example the Stiff diagrams presented in Plate 2 (EM,1995) show bicarbonate concentrations of 20 meqA (eqwvalent to 1200 mg/l)
The Lower Hydrostratigraphc Unit exhibits lower bicarbonate levels, as shown m Figure 6-46 (EG&G, 1999, where a uniform concentration of 100 - 200 mgll is observed site wde, apart from area around the Industnal Area, where concentrations rise to above 200 mgA
Sulphate
Sulphate is an important anion producing enhanced degradation of concrete Sulphate is also a potential microbial substrate, which could enhance concrete degradation and radionuclide mobility Sulphate is generally low in the UHSU, typically less than 10 mg/l in the Woman Creek Flow Path background wells In the Industnal area sulphate is much more vanable increasing to over 1000 mgA The groundwater around the Solar Evaporation Ponds and Ponds B along the Walnut Creek appear to be particularly enhanced wth respect to sulphate concentrations
Figure 6-5 1 (EG&G, 1995) indicates that the concentration of sulphate in the LHSU is farly constant across the site, less than 100 mgA Thls may indicate that the background concentrations of sulphate are slightly hgher in the LHSU than in the UHSU There are increases in the Industnal Area, but these do not appear to be as high as those in the UHSU
Chloride
Chloride produces enhanced degradation of concrete and induces corrosion of steel reinforcements (rebars) Chloride generally behaves like the other major elements, as concentration tends to increase along the flow paths In the UHSU, background concentrations are less that 25 mg/l, whle the concentrations at the WETS plant are generally between 25 and 50 mg/l, wth hot spots, again, around the industrial area (where concentrations exceed 500 mg/l in places) In the LHSU, concentration is fairly uniform at less than 50 mg/l
Reduced Nitrogen, Sulphur and Organic Matter
Reduced species are required for microbial degradation, and nitrogen, sulphur and carbon are often present as reduced species in soils and groundwater There is no mention of reduced nitrogen, in the form of ammonium, in any of the groundwater analyses, which is reflected in the redox measurements (see later) Pynte is present
Page 10 of 57 Runed Concrete at WETS-A Literature Review in cemn areas of the site, and th~swould appear to be only source of sulplude Orgmc composition is less clear, and apart from anthropogenic orgmc compounds, the presence of orgmcs is not mentioned
pH is an important control on radionuclide mobility, and on alkali and Ca leaching of concrete The background pH at the WETS appears to be between 7 5 and 8, and appears to be farly constant along all four flow paths, no doubt due to calcite equilibnum The highest pH’s occur around the Industnal Area, where the pH reaches values above 8 Groundwater close to the Landfill exhlbits relatively low pH’s (below 7) In general, there is relatively little pH vanation over the site, in the UHSU The isoconcentration map for the LHSU does indicate three areas where the pH is above 10 However, these appear to be isolated and the pH is otherwise below 8
Redox
The oxidation state of redox sensitive elements such as plutomum and uranium is important to their mobility Redox measurements can also give an indication of the extent of microbial activity The amount of redox data avalable from the site is sparse, and the difficulties in measuring Eh are acknowledged The redox potentials were measured using a flow-through cell, and the resulting Eh’s are in the range of 0 09 to 0 32 V i e mildly oxidising The presence of sidente and pyrite suggests reducing conditions, thus either there are pockets of anaerobic waters or, more likely, the presence of these minerals is indicative of the conditions in the past when these minerals were formed It is mentioned that potentially low Eh’s may be encountered close to the Landfill, but the values are not presented in the report
Cement and Concrete Degradation
Introduction
Several physical, chemical and biological processes contnbute to the degradation of cement and concrete (Lagerblad and Tragardh, 1996, Glasser, 1997, Rogers, 1993a) Many of the processes are inter dependant and involve some aspect of chemical degradation Chemical degradation reactions of cement minerals are important as they control the local pore fluid chemistry where radionuclide contamination may exist Fluld chemistry exerts a strong influence on the solubility of the actinides and in some cases radionuclides may co- precipitate with cement alteration products Sorption of radionuclides is also influenced by fluid chemical conditions such as pH as well as the surface mineralogy It is therefore necessary to review the major chemical reactions involved in cement and concrete degradation
The main chemical processes affecting cement and concrete are 0 Leaching and acid attack Carbonation 0 Sulphate, magnesium and chlonde attack 0 Alkali-aggregate reactions 0 Corrosion of steel rebars
Page 11 of 57 @BNRInc - - Bund Concrete at WETS-A Literab~reReview --=-m Biodetenoation of concrete pnncipally involving microbial induced attack facilitates localised chemical degradation by several processes leadmg to enhanced surface degradation of concrete Since contamination in WETS concrete is concentrated at the surface such processes operating at the groundwater interface must be considered to be potentially important in releasing radionuclides
The mobility of radionuclides in groundwater can be enhanced by the presence of suspended particulate colloidal materials onto which radionuclides may be sorbed Since concrete degradation may produce such fine-gramed matenal avadable literature on colloid generation is considered
An understanding of long-term degradation of concrete can be obtamed fiom study of ancient analogues of modern concretes and fiom computer models Computer models include prediction of the structural evolution of concrete repositones and the prediction of chemical degradation and pH evoluhon Analogues and models provide an estimate of the time-scale of concrete degradation and the likely extent of degradation dmng the 1000-year assessment penod
Followng review of the above aspects of cement and concrete degradation the processes most relevant to WETS w11 be discussed wth consideration to the nature of radionuclide contammation and site-specific groundwater conditions
Chemical evolution of cement minerals
Hydration of Portland Cement (OPC) leads to the formation of a vanety of chemical phases which make up hardened cement (Lea, 1980) The initial pore solutions formed on dissolving anhydrous OPC are strongly oversaturated and hydrated silicate phases form in eqwlibnum wth the pore soluhon The kinetics of hydrated OPC mineral formation is however slow, imtial phases formed are amorphous gels whch form more crystalline phases wth accelerated ageing, and at increased temperature The compositions of the initial gel hydration phases formed are vanable and the sequence in whch they form is partly dependent on temperature and cmng conditions Dung leachmg by pure water, alkalis and calcium are removed fiom the cement minerals, which results in a decrease in pH of the cement pore fluid and change in composition of the cement matnx .
Hydrated cement minerals
A standard nomenclature is used to abbreviate OPC and hydrated cement mineral formulae in most publications on cement and concrete
Abbreviation Oxide Formula C CaO S sio, A A1203 F Fez03 S so3 H H2O
Page 12 of 57
*, ~BNFL Buned Concrete at WETS-A Literature Rcvlcw - Inc. The symbols m and t are commonly used to denote the ratio of calcium in a mineral and correspond to mono- and trl- respectively e g monosulphate AFm
Some of the man hydration products are Calcium-silicate-hydrate (CSH) - the man component of hydrated OPC Amorphous to semi-crystalline wth CdSi molar ratio of 0 8 to 3 0, wlth vanable water content CSH forms dmng the hydration of anhydrous trl- or di- calcium silicate wth water yielding Ca(OH), and a more silica nch CSH Calcium hydroxide - a crystalline product forming large crystals Aluminium-iron-mono (AFm) these phases have the general formula (Cq(Al, Fe) (OH),xXyH,O, where X is an equivalent ofa single charged anion Monosulphate and Fnedels's salt are important hydrated cement phases contaming sulphate and chlonde respectively Tn-calcium-alurmnates (AFt) formed by hydration of calcium aluminate, these phases can contam mons such as SO,- and C0,- The sulphate mineral ettnngite is an important member of th~sgroup, resulting in sulphate promoted degradation Hydrogarnet - a solid soluhon series contamng Ca, Si and Al, whch coexist wth CSH phases Brucite (Mg(OH),) small amounts of this phase are present in hydrated cement Secondary brucite is responsible for Mg promoted attack
In pure OPC based concretes wth low concentrations of reactive alumina, calcium hydroxide and CSH are the man phases present Alumina contaming phases are more important to applications of cement grouts and backfills used in the encapsulatlon of radioactive waste which include blast furnace slag and pulvensed fly ash as additives, such materials contain a high content of pozzolanic reactive alumina which results in the formation of aluminous hydrated cement phases Aggregate added to concrete may contain alumina in the form ofrock forming minerals, however, these phases are largely unreactive
Leaching of Ca(OH), and CSH
Extensive expenmental studies have been performed examinmg the leachng of Ca(OH), and CSH in pure water (Berner, 1987, Adenot et al, 1992, Atluns et al, 1992a,b,c, 1994, Engkvist et al, 1996, Delagrave et al, 1997, Glasser, 1997, Bennett et al, 1992, Pfingsten and Shotsulu, 1998, Duerden et al, 1997, Nedl, 1996, Qwllin et al, 1994) The majority of these studies have been performed to examine the pH behaviour dmng leaching which is an important control on the mobility of radionuclides fiom cement-based grouts and backfill matenals The leaching ofOPC based cements can be divided into five main periods (Figure 1)
Flushing of residual alkalis (NaOH and KOH) from pore spaces at pH 12-14 Leaching in the presence of Ca(OH), pH is buffered at over 12 0, CSH co-existing wth Ca(OH), has a CdSi ratio of approximately 1 8 When Ca(OH), is exhausted pH is controlled by the CSH phase which dissolves incongruently, preferentially releasing Ca into solution pH decreases from over 12 0 to -10 5 as CSH changes in composition from CdSi ratio 1 8 to 0 8 On reaching a CdSi ratio of -0 8 CSH dissolves congruently and pH remains constant at -10 5 until the CSH finally dissolves Following dissolution of CSH pH is controlled by secondary Ca minerals, typically calcite under groundwater conditions pH drops to - 7-8 dependent on carbonate levels in the groundwater
Page 13 of 57 - - Bmed Concrete at WETS-A Literature Review The durabon of these pH buffers depends on the amount of each mmeral in the concrete, the permeability (related to the quality of the concrete), and the groundwater chemcal composition The effect of alkalis is relatively short-lived since these are present m the initial porespace and are removed after flushing of 2-3 pore volumes The amount of free Ca(OH), is determined by the amount of pozzalan~cmatenal present in the concrete, excess Ca(OH), reacts wth such matenal forming CSH and CASH phases Pure OPC concretes used for construction purposes, which do not contain blast fiunace slag or pulvensed fuel ash w11 conan sigmficant amounts of free Ca(OH), CSH form the man buffenng phase, the rate and duration of dissolution of Ca depend on temperature, surface area and the concentration of Ca, Si and carbonate in the groundwater Dissolution is most rapid in soft CO, -nch water, and slower in CaCO, waters Acidic water produced by microbial action, such as sulphde oxidation, or acidic spillage onto concrete w11 effectively accelerate the leaching process Under accelerated acidic leaching diffusion processes limit the leaching process (Lefebvre, 1997) The sequence of changes in cement mineralogy occurnng dung leaching illustrated in Figure 1 can be considered to occur either as function of time, or fluid volume at a given point, or as function of distance representing zoned alteration of a concrete surface
14 Alkalis KOH/NaOH 13 Ca(I)"l 1 CSH I Incongruent CSH dissolution 12
11 \CY Silica , % 10
9
8 1 ------Ca leaching Carbonation / 7 Time, leach volume, extent of degradation
v Depth from surface
Figure 1 Evolution of cement Dore fluid pH dwng leachng Page I4 of 57 @BNFLInr Buncd Concrete at WETS-A Lentun Rcvlew
Carbonation
Carbon dioxide reacts strongly wth the alkaline components of OPC concrete dmng the process of carbonation Both Ca(OH), and CSH react wth CO, to form calcite
Ca(OH), (s) + CO, (g) = CaCO,(s) + H2O
CSH + CO, (g) = CaCO,,, + SiO,,, + H,O
Carbonation can occur in both saturated groundwater, in the unsaturated zone, and in above ground structures In soils carbonation is favoured by high CO, content and the humidity of the soil gas Optimum humidities are around 75%RH (Houst,1997), the carbonation reactions only occur m the presence of water, however water produced by carbonation must be allowed to diffise out of the carbonated layer Cement permeability and onginal waterkement ratio are therefore important to the extent of carbonation Tuum (1 982) has invesbgated the effects of permeability on carbonation (Figure 2), in ramwater a good quality concrete wth W/C 0 45 w11 carbonate to a depth of 5 mm in 50 years In groundwater where CO, levels are higher than in the atmosphere, carbonation wll be more effective (Lagerblad & Tragardh, 1996)
Page 15of57 @BNFLInc . BwedConcrete at WETS-A Literature Review
100 Not sheltered from rain 0 4 - - - Sheltered from rain 0 50 0 / 0 0 0 0 30
E 20 E waterkement ei Um 5 075 0 3 10 *0 0 5n d 5
3
2
I I 3 I - 2 345 10 20 30 50 100 Time, years . Figure 2 Measured mean carbonation det>thin Portland cement concrete wth vming waterkement ratios After Tuutti, (1982). LaPerblad & Tragardh, (1996)
Formation of calcite from Ca(OH), results in an increase in volume of 12% (Houst, 1997) which reduces porosity Despite a decrease in porosity overall shrinkage occurs (Houst, 1997) and microcracks develop (Walton et al, 1997) Decreases in porosity are beneficial in view of degradation by freeze-thaw mechanisms Overall the durability of cement increases dung carbonation Carbonation results in a large decrease in pH to -8, where steel reinforcement is more prone to corrosion and where microbial induced degradation may become established Studies of contaminant leaching from carbonated cements show vaned behaviour, some contaminants such as Sr are incorporated in the secondary calcite (Walton, et al, 1997, Curti,l998) while unreactive species may show increased diffusion as a consequence of microcrack development (Walton et al, 1997), or lower difision in high w/c cements (Sarott et al, 1992)
Page I6 of 57 @BNFLInc Buried Concrete at WETS-A Literahue Revlew
Sulphate, Magnesium and Chloride attack
Accelerated degradation of hardened OPC concrete occurs by the formahon of secondary sulphate phases, by reaction of hydrated cement phases with groundwater sulphate Formation of sulphate minerals results in a volume increase, whch eventually produce expansion, and may result in fracturrng and spallation of concrete Ettnngite is the sulphate phase most commonly associated wth sulphate attack and forms by reaction of calcium aluminates and monosulphate (C,A CaSO, 12H20) in OPC Ettnngite is a pnmary component of hydrated cement resulting from the inclusion of gypsum to control setting time Secondary ettnngite forms by further reaction of sulphate in groundwater Sulphate-resisting cements are formulated which contam a low proportion of alumina to reduce ettnngite formation in sulphate waters At increased sulphate concentrabons, gypsum (CaSO, H,O) forms by reaction of Ca(OH),, in 5% sulphate solution ettnngite and gypsum form concurrently (Ferrans et a1 1997a) Expansion of cement exposed to sulphate solut~onsoccurs imbally by a diffusion controlled mechamsm until avadable porespace is filled by gypsum and ettnngite, crackmg then follows whch allows further access of sulphate soluhon to unreacted monosulphate (Pommersheim and Clifton, 1994)
In the presence of both sulphate and carbonate the mineral thaumasite (CaSiO, CaCO, CaSO, 15H20)can form from alteration of CSH phases Since CSH is the man binding agent in the cement hsform of sulphate attack results in complete breakdown to a dissaggregated mush (Crammond and Halliwell, 1997) Thaumasite attack is limited in occurrence and is favoured by low temperatures and use of limestone and in particular dolomite aggregate
Magnesium sulphate solution is more deleterious to concrete than alkali sulphates because CSH phases are attacked as well as the aluminous phases Under the hgh pH conditions of the cement pore fluid brucite (Mg(OH),) precipitates and lowers pH, so that the CSH is destabilised, CSH then forms gypsum and free silica as follows
3Ca0 2S10, + MgSO, 7H,O = CaSO, 2H,O + 3 Mg(OH), + 2 SiO,
There is a net expansion dmng this reaction which results in hctunng, the formation of brucite produces a hard skin on the surface of mortar and concrete and this can restnct Merattack
The extent to which sulphate attack will take place wll depend on the sulphate concentration of percolating groundwater Secondary ettnngite will only form if the groundwater has a hgher sulphate concentration than that of the cement equilibrated pore solution The sulphate concentration of the cement porewater is controlled by temperature and alkali content (Lagerblad & Tragardh, 1996) and can vary between 0 1 and 100 mM/1 SO," (Damidot et a2 1992) No precise groundwater SO,' concentration can be specified at which sulphate attack becomes significant A significant loss of compressive strength occurs after one years storage in 5% sulphate (Lea, 1980), while at a concentration of 0 5% sulphate magnesium sulphate attack is significant but sodium sulphate has little effect Atkinson and Hearne (1990) have developed a mechanistic model to predict the long- term durability of concrete exposed to sulphate groundwater
Chlonde ions in groundwater can have a similar effect on concrete degradation as sulphate Normal cements have very low chlonde contents to avoid corrosion of steel reinforcement Chlonde ions bind to the AFm phase and produce Friedel's salt 3Ca0 Al,O, CaCI, 10H20,at very high concentrations of chlonde the phase tnchlonde 3CaO A120, 3CaC1, 32H20 is formed The action of chlonde may also produce reachons in the Page 17 of 57 ~BNFL - - Buried Concrete at WETS-A Literature Review Inc sulphate cement phases (Lagerblad & Tragardh, 1996) where chlonde replaces sulphate in monosulphate whch then is avadable to produce ettrrngite Reactions among chlonde and sulphate cement phases are dependant both on temperature and chlonde content The AFm chlonde phase is stable between 10 and 1000 mmoVl C1 at 25OC (Atkins et al, 1994) Chlonde concentrations of around 300 mmovl and temperatures above 50°C are required to produce decomposition of ettrrngite (Lagerblad & Tragardh, 1996)
Alkali - aggregate reactions
Quartz and other rock forming minerals are unstable under the strongly alkaline conditions produced by the presence of sodium and potassium hydroxide in concrete pore solutions Crystalline minerals such as quartz are however slow to react even under strongly alkaline conditions More reactive forms of silica such as opal, chalcedony and glassy acid to intermediate volcmc rocks are liable to react to form an alkali-silica gel and CSH Formation of these phases as alteration rims around aggregates produces expansive forces, whch may result in cracking (Ferrans et al, 1997b) and exhudation of a soft viscous gel (Lea, 1980) Such alkali- aggregate reactions are dependent on the alkali content (NqO + K,O), below 06% NqO usually no deleterious alkali-silica reactions occur over short time-scales (1 00 years) (Lagerblad & Tragardh, 1996) Formation of alkali -silica gels removes alkali and lowers pH and thls induces dissolution of Ca(OH),, released Ca may then exchange wth alkali-silica gel to release alkalis and induce further alkali-silica reaction (Wang and Gillot 199 1, Lagerblad & Tragardh, 1996) The rate of alkali-aggregate reactions is controlled by the rate of diffusion of alkalis to the reaction site wthin the aggregate or on the aggregate surface and by the dissolution rate of silicates Silicate dissolution rates are strongly pH dependent and increased at hgh pH (Lasaga, 1984) Contact of groundwater with concrete wll clearly reduce the susceptibility for alkali- aggregate reactions by removing alkalis and lowering pH
Corrosion of steel reinforcements
Corrosion of steel reinforcement (rebars) results in expansion and ultimate fkactunng of concrete Formation of FeO from Fe results in doubling of volume, whde the formahon offemc hydroxide (Fe(OH), 3H,O ) increases volume by a factor of 6 5 (Li and Li, 1997) Steel corrosion is thought to be limited under the hgh pH conditions due to the formation of a passivating protective surface (Wheat et al, 1997) Corrosion is induced in concrete by the effects of chlonde and carbonation Chlonde-induced corrosion is generally considered more important (Li and Li, 1997, Constantinou and Scrivener, 1997), largely because the effects of carbonation are slow
A number of mechanisms are likely to be responsible for chlonde-induced corrosion (summansed in Wheat et al, 1997), chloride may attack the protective fildsubstrate bond without attacking the passivating layer or the layer may be chemically attacked Other theones propose that chloride is preferentially adsorbed in competition with dissolved oxygen and hydroxyl ions, or that chloride ions may penetrate the oxide film more easily than other ions Once the passivating layer is broken galvanic corrosion occurs Steel corrosion rapidly consumes dissolved oxygen and then produces hydrogen and establishes reducing conditions
Carbonation-induced corrosion proceeds by lowenng the pH of the concrete pore fluid below that at which a passivating layer forms Although the rate of carbonation is affected by the quality of the concrete
Page I8 of 57 @BNFLInc Burred Concrete at WETS-A Litentun Revlew (waterhement ratio), once corrosion is imtiated the mitial properties of the concrete have no effect (Constantinou and Scnvener, 1997)
Biodeterioration of Concrete
Biodetenoration can be defined as any undesirable changes in the properties of a matenal by the activity of organisms i e plants, animals and micro-orgamsms In the specific case of concrete the orgamsms of interest are on the whole micro-orgamsms, and these organisms wll be the focus of thls section Micro-orgasms are ubiquitous in all natural and man made environments and have been associated wth the detenoration of many commercially important matenals including construction matenals such as steel, stone and concrete This chapter wll descnbe the processes by whch microbial biodetenoration of radioactwely contaminated concrete can take place, present examples from other industnes where appropnate and outline the factors which influence the rate and extent of biodetenoration
Microbially Induced Degradation of Concrete and Cement
Microbially induced degradation of cement based matenals or MID has been extensively studied in the nuclear industry due to its potential role in the detenoration of containment in cemented waste forms (Rogers et al 1993a,b, 1994, 1996) and has even been developed into a biologically based decontamination process for surface contaminated concrete (Rogers et al, 1997) In non-nuclear industnes MID has been associated with the degradation of sewage pipes, water distribution systems, power station cooling towers and buildings (Rogers 1993a)
MID has two mam aspects, firstly the deterioration of the concrete surface by the action of micro-organisms resulting in a loss of structural integnty This results in the surface of the concrete being susceptible to sloughing off carrying any associated radionuclides wth it Secondly micro-orgamsms have the ability to secrete complexing agent whch may mobilise radionuclides which are weakly bound to the concrete Once complexed in this manner these radionuclides are susceptible to further mobilisabon via water flow for example
Direct deterioration of the concrete surface occurs when micro-organisms growng on the surface of the concrete generate acidic compounds whch attack the concrete in the same manner as chemical acid attack The fact that in MID the process is microbially mediated usually results in the impact being more severe than direct chemical attack This is because the micro-organisms grow in intimate contact wth the surface and consequently any acid generated impacts on the surface as a concentrated point source There are three well- known types of microbially mediated acid attack which are outlined in figure 3 The pnmary mode of acid attack is one of enhanced leaching where the inherent alkalinity of cement pore fluid is neutralised and alkalis and calcium are removed as soluble salts The specific soluble salts formed depend on the specific acid involved in the attack
Page 19 of 57 @BNFL .) - Buried Concrete at WETS-A Literatun Review Inc
Figure 3 Tv~esof Microbially Mediated Acid Attack
The generation of sulphunc acid is catalysed by sulphur oxidising bactena wth bactena from the genus Thzobaczllz being most commonly associated with concrete degradation Thzobaczllz species get their energy for growth from the oxidation of reduced sulphur compounds wth the subsequent generation of sulphmc acid The majonty of these bactena use molecular oxygen to drive this oxidation but there are species which can use nitrate in the absence of oxygen This group of bactena has a wde tolerance to acidity wth Thzobaczllz species growng at pH’s ranging from pH 6 5 to below pH 4 0 (Smith and Strohl 1991) The bactena deposit sulphuric acid directly on the surface of the concrete as they grow Since sulphuric acid will react wth and destroys Portland cement at any temperature above freezing (Hall 1989) this form of MID can have a devastating effect on any concrete structure under attack The result of sulphuric acid attack is a loss of granular structure and the formation of sulphate salts such as gypsum (Rogers et a1 1993a) Sulphate may also be avalable for attack of AFm phases forming ettnngite
In order for Thzobacrllz species to grow they require sources of oxygen, carbon dioxide and reduced sulphur In nature the reduced sulphur source is generally found in the form of pynte Pynte oxidation is commonly associated wth a specific species of Thzobaczllz, Thzobaczllus ferrooxzdans This bactena can oxidise both reduced iron and sulphide to generate energy for growth, and is the key species in the generation of acid mine dramage and microbial mining of metals such as copper and nickel (Bnerley and Brierley 1997) In other cases Thzobaczllz species may grow on reduced sulphur compounds generated by another group of bactena, the sulphate reducing bacteria (SRB) Th~sis important since it allows sulphuric acid based MID to occur where no obvious source of reduced sulphur exists In thls situation the sulphate reducing bacteria reduce sulphate to generate hydrogen sulphide as part of normal growth This process occurs under strong reducing conditions, the sulphide generated then migrates into an oxidising environment where Thzobaczllz species are able to utilise it generating sulphuric acid as a by-product The classic example of this cycling of sulphur resulting in concrete degradation is in sewer systems Here, SRB’s are able to grow in the liquid waste in the bottom of the sewer pipes generating hydrogen sulphide This migrates as a gas into the body of the sewer pipe wth some of it dissolving into condensation on the concrete surface Here, where there is both a source of oxygen and sulphide Throbaczllz are able to generate sulphmc acid This mechanism has been responsible for catastrophc falure of the sewer system in Hamburg, Germany (Rogers et al1993a) Page 20 of 57 @BNFLlnc Bund Concrete at WETS-A Literature RCVICW
Nitnc acid attack is similar to sulphmc acid attack m that a bactena catalyses the generation of acid through the oxidation of a reduced compound dmng its growth Thls is classically caused by two groups of bactena collectively known as nitnfyers The first group oxidises ammonia to mtnte and the second nitnte to nitrate Both reactions liberate hydrogen ions and generate nitrous and mtnc acids respectively The bactena responsible for these reactions are common in soils and aquatic environments such as nver beds Unlike the sulphur oxidising bactena, the mtnfyers and specifically those responsible for ammonia oxidation, are sensitive to low pH’s Thls means that nitrogen base MID is self-regulating and does not generate the low pH’s associated wth the growth of Thzobaczllz species
Nitnfyers have been isolated fkom corroded concrete and implicated in the corrosion of stone with acid depleting the binding matenal (Rogers et al 1993a) A classic example of mtnc acid MID has occurred on Cologne cathedral and other sandstone buildings in Germany (Rogers et a1 1993a)
The bacteria responsible for the generation of organic acids are a much more diverse group than those discussed up to now in this section These heterotrophc bactena generate energy for growth through the consumption of complex orgmc compounds and may generate a wde vanety of orgmc acids as by-products of their growth These acids include acetic, lactic, citnc, gluconic etc, all of which may attack cement-based matenals The wide vanety of orgmsms, which fall into this category, means that they are very wdely distnbuted in all environments, particularly soil The role of heterotrophic bactena in MID is not as extensively studied as that of the sulphur oxidisers or the nitrifying bactena However, they have been isolated from concrete corrosion sites and it is suggested that they play a role in reducing the pH to a value, which is more favourable for the growth of Thzobaczllz species (Rogers 1993a)
Generation of Complexing Agents
From the point of view of radioactively contaminated concrete, organic acid attack has an additional effect whch may be important This is the fact that these acids and particularly citnc acid, can complex radionuclides resulting in an increased solubility In the case of citrate plant roots and fhgal hyphae have been implicated in its release into soils (Wlute et a1 1997) The degree to which complexation is a factor wll depend on the manner in whch the radionuclides are immobilised on the concrete surface It is possible that the action of direct acid attack and complexation act in concert With the acid attack resulting in the radionuclides becoming more susceptible to complexation since their attachment to the concrete is weaker
In addition to the organic acids, micro-orgmsms are known to produce a whole range of organic molecules which can complex metals These include siderophores which are specifically excreted under iron limiting conditions, and metal binding proteins such as metallothioneins The extent to which these compounds are important in the mobilisation of radionuclides is difficult to assess and their generation may be a source of significant uncertainty However, 2-ketogluconic acid has been implicated during the weathering of silicates (Webley et a1 1963) and oxalates dmng basalt weathering (Silverman and Munoz 1970) Complexation by these compounds cannot therefore be neglected in consideration of risks
Page 21 of 57 @BNFLInc. --Buried Concrete at WETS-A Literature Review
Factors Influencing Biodetenoration
In order to assess the impact of biodetenoration on concrete and the release of radionuclides it is important to understand the factors wluch control biodetenoration Although the surfaces on whch micro-orgasms are growng dwgMID are generally inert the orgmsms must be in contact wth an environment whch provides them wth the nutnents they require to grow The major controlling factors for biodetenorahon are outlined below
Water avai LabiLity
Water is essential for microbial growth, consequently wthout an adequate amount of water MID w11 not proceed Thls water may be provided as a liqud or a vapour and could come fiom the smundmg soil for a buried structure or fiom humidity for a above ground structure The influence of water on the extent of MID is such that it may proceed intermittently in environments where there are pronounced wet and dry seasons
Water logging can also have a marked effect on the activity and types of micro-orgasms present ~flsoil 'Ths is because under these conditions oxygen generally becomes limited and anaerobic conditions prevail This would promote the generation of organic acids such as acetic acid but retard Thzobacrllr species and nitnflers
The pH can have a significant effect on the growth of micro-organisms wth each species having a specific range and optimum under whch it grows As we have already discussed the pH range for ThzobacrZlz species and therefore sulphmc acid attack is very wide, where as that for nitnflers and nitnc acid attack is narrower In the case of concrete degradation the surface pH may have to be lowered through carbonation for example, before ThzobaczZZz species can take hold Thls lowenng of the imtial pH may also be aclueved by the presence of heterotrophic micro-orgasms generating orgmc acids
Nutrients
The availability of nutnents is key to the progress of MID If there are no reduced sulphur sources or reduced nitrogen source avilable then MID via Thzobacrllr species or rutriflers w11 not proceed On the other hand if there is insufficient oxygen, then the availability of these sulphur or nitrogen compounds w11 not be important since the system will be oxygen limited In terms of heterotrophic generation of organic acids the key will be the availability of organic substrates in the soil to drive the generation of the acids The most important variable in assessing the likelihood of the vanous types of MID is therefore the availability and quantity of the particular nutnents required to dnve the process
The situation with the complexing agents may be more complex Thls is because some of these compound are generated by micro-organisms in response to nutnent shortages and in effect represent a survival strategy @BNRIM Burred Concrete at WETS-A Literature Rcvtcw
Temperature
As with pH micro-orgasms have a optimum and a range of temperatures over which they grow Under environmental conditions the general trend is as the temperature increases so does microbial activity As wth water avsulability hsmay have the effect of producing seasonal vanations in the extent and rate of MID
Presence of Micro-organisms
Generally the presence of the partmlar microbes associated wlth MID is not a controlling factor This is because all the orgasms rnvolved are ubiquitous in all natural and man made environments In many cases they have the ability to stay dormant for long penods of time until the conditions whch are favourable for their growth become available
Colloid Generation in Cementitious Systems
Very fine-grained suspended solid matenal can potentially act as a separate transport mechasm for radionuclides to that of dissolved species transport Plutonium and other actimdes may form aggregates of polymenc aqueous species (true colloids) and these are discussed in the section on speciation Radionuclides may also be transported as sorbed material onto inorganic(minera1) and organic particulates (pseudocolloids) Such material is typically in the size range lnm - 1pm (Kim et al, 1997) A number of studies of natural groundwaters including analogues of actimde contamination have shown that a sigmficant proportion of radioactivity can be present in groundwater present on such pseudocolloids (Olofsson et al, 1986, Kim et al, 1987, 1989,1997,Longworth et al, 1989, Miekeley et al, 1989, Dearlove et al, 1989) Colloidal matenal which has been shown to sorb actinides includes, organscs (humics), iron hydroxides, clay minerals, silica, and other silicate minerals (feldspar)
Direct investigation and charactensation of colloid formation in cemenbtious systems has been made by Ramsey et al, (1988)(see also Gardiner et al, 1997) Colloids formed in leachates from OPC were filtered to collect vanous size fractions and were examined by electron microscopy, X-ray diffraction and bulk chemical analysis (ICP-AES, ICP-MS) Colloids collected were composed of CSH and effectively represent fine - grained cement matnx Colloids produced showed significant sorption of radionuclides including uranium Leaching of OPC in these expenments was performed in closed flasks purged with nitrogen, thus eliminating carbonation, it would be expected that fine-grained suspended CSH matenal would undergo carbonation in an open system wth freely available CO, Dmng carbonation CSH produces amorphous silica which may form colloidal matenal
In order to increase radionuclide mobility sigmficantly sufficient colloid particles must be present in suspension Gardiner et ul, 1997 have considered the processes controlling colloid particle growth and aggregation, nucleation, temperature and ionic strength are factors which govern the colloid concentrahon in expenments on colloid generation from OPC leachates Gardiner et al comment that when included in the Nirex 97 nsk assessment of the post-closure radiological safety case of a U K repository cement generated colloids did not significantly increase nsk Bradbury and Sarott (1994) also comment on the ability of
Page 23 of 57 @BNFLInc. -.. Buned Concrete at WETS-A Literature Review
cementitious colloids to sigmficantly increase the total concentraQonof radionuclides in the aqueous phase of the near-field repository environment, since CSH is the major component of both colloids and cement solid phase sorption propemes w11 be simlar Bradbury and Sarott (1994) conclude that unrealistic amounts of colloidal matenal (> 10 gramditre) wll be necessary to significantly increase aqueous phase radionuclide concentration Similar arguments may apply to the generation ofcolloids from carbonated concrete if it can be demonstrated that the solid substrate and colloid matenal are composed of similar matenals
Ancient analogues of concrete degradation
Ancient analogues of modern OPC based concretes have been examined to provide information on the long- term behaviour of concrete and mortar in radioactive waste repositones (Lagerblad & Tragardh, 1996) The greatest difficulty in applying such information is the vanation in the composition and gram size of ancient C,S cements Portland cement used in the early 20th century were coarser and contamed a higher content of than that used today OPC based cements exposed to water for penods of the order of 100 years remam durable, there is evidence that cement phases requilibrate and form larger crptals, and in some cases there is evidence of reaction wth cement aggregates Avzulable examples show relaturely little carbonation when immersed in water (5mm in 90 years, Lagerblad & Tragardh, 1996) More ancient Roman mortars and concrete over 2,000 years old which contamed pozzolmc matenal resembling OPC type cements still show strong durability (Jiang and Roy, 1994) These cements are hlly carbonated and contam only a small amount of calcium silicates (Majumdar et al, 1988) Carbonation occurs in these ancient buildings because of exposure to the atmosphere (Lagerblad & Tragardh, 1996) Modern pollution, in particular acid mn, has had a marked influence on the durability of some ancient monuments such as the Taj Mahal and the Pyramids (Roy and Jiang, 1997) and indicates the importance of modern sulphate and nitrates to concrete degradation
Repository degradation models
Long-term models have been devised to predict the structural and chemical evolution of concrete structures for low and intermediate level radioactive waste disposal Models have been proposed of varying deml, and considenng vmous chemical and physical processes of cement and concrete degradation Lagerblad & Tragardh (1996) have produced a conceptual model for the chemcal and structural evolution of a proposed Swedish deep nuclear waste facility This model identifies the processes and estimates the extent of chemcal degradation during the penods of site development and post-closure, where diffenng groundwater, headspace gas composition and temperature conditions occur For this deep repositorj extremely long time-scales, of more than 100,000 years, are considered and only a qualitative descnption of the behaviour of concrete is provided Mathematical models have been developed to quantitatively predict the service life of concrete buned near to the surface over penods up to 1000 years used to store low-level radioactive waste (Atkinson and Heme, 1989, Reed et al, 1994 report, Snyder et al, 1996, Gerard et al, 1997, Lee et al, 1995) Most of these models utilise ernpincal analytical equations to model individual degradation processes such as sulphate attack, Ca(OH), leaching, and carbonation Corrosion of steel rebar can be modelled by considering the diffusion of chloride to the steel and from an anoxic corrosion rate The model of Reed et a2 (1 994) links these degradation processes to a structural analysis code which computes stresses in concrete and weakened rebar at vmous locations in a structural model Once stress in the rebar exceeds a certam threshold cracks develop in the concrete as a function of the structural design, at a higher threshold the rebars yield and additional stress is applied to other parts of the structure The model successively calls the concrete degradation and structural sub-models untd some predefined state ofcollapse ofthe structure is reached Snyder et a2 (1 996) consider the
Page 24 of 57 @BNRInc Burred Concrete at WETS-A Literature Review advective and diffusive transport of ions through a repository concrete slab and its associated degradahon by sulphate attack, steel corrosion and Ca(OH), leachmg Gerard et aZ (1997) consider simultaneously the specific interactions between diffusion, leachmg, mechmcal strength, craclung and permeahon Alcorn et a2 (1 990) have denved a model to predict the hydraulic conductivity of OPC cement grout to be used as a cement backfill matenal, according to this model for an ambient hydraulic head of ldmthe hydraulic conductivity remans below acceptable performance level (lo-'' dsec) for a minimum penod of 30,000 years Adenot and Rxhet (1997) descnbe a model of purely diffusive reaction of water wth cement paste whlch incorporates the chemical leaclung behaviour of CSH and the breakdown of monosulphate and ettnngite According to Adenot and hchet (1997) for a good quality cement (waterhement = 0 4) a degradation layer composed of CSH and silica gel w11 extend a hckness of 1 2 mrn der 3 months and 4 cm after 300 years Purely diffusive alteration of cement is limited by the build up of a protective layer of silica gel, if th~sgel is removed by some other process such as erosion or chemical dissolution then degradation is more severe
Concrete degradation relevant to radionuclide mobility at Rocky Flats
Concrete degradation effectively controls release of actimde contamination by allowng free access to moving groundwater Concrete has a low permeability compared to soil matenal and therefore it is unlikely that there will be significant water flow through concrete blocks bmed wth soil, flow will concentrate around the outside of concrete blocks The bulk of the contamination in WETS concrete is present in the upper few millimetres of concrete The mobility of radionuclide contamination wll therefore be pnmmly controlled by the zoned chemical degradation formed on the surface of the concrete The nature of the degradation w11 clearly depend on the groundwater environment and geochemistry
The background groundwaters at RFETS wthin the upper hydrostratigrapluc mt are typical dilute Ca bicarbonate waters, whereas groundwater at depth in the lower unit are more sodium rich and more vanable The upper hydrostratigraphic unit is vanable in thickness from 10 ft to 130 ft therefore it is likely that bmed concrete will be in contact wth a dilute Ca bicarbonate water Such water is unlikely to produce enhanced degradation of AFm phases in the cement matnx by attack by sulphate, chloride or magnesium In the mdustnal area however, there are some hgh concentrations of sulphate over 1000 mg/l, these concentrations are likely to produced enhanced degradation by formation of secondary ettrrngite Concrete burred in the unsaturated mne is likely to be subjected to carbonation CO, partial pressures in soil gases are typically above that of the atmosphere as a result of respiration by plants and micro-organisms, and thus carbonation will probably be accelerated dmng burral Pmal pressures of CO, calculated from speciation calculations are not summarised in the RFETS Site Charactensation Report (EG&G, 1995) The example WATEQF output in Appendix H of the Groundwater Geochemistry Report however gives a CO, partial pressure of 6 93e-2 atm which is above the maximum of CO, levels normally measured in soils (log pC0, = -1 5 Appelo and Postma, 1994 ) While this value may not be typical CO, contents are likely to be significantly above that of the atmosphere, and w11 thus promote carbonation Fujiwara et aZ(1992) examined concretes burred in soil for 60 years in saline groundwater, they observed that calcite was the main alteration product in the upper 10 cm of the concrete, and that chlonde alteration occurred at greater depth as a consequence of the saline water The samples examined by Fujiwara et aZ were in saturated water where carbonation is generally thought to be less effective More ancient Romanic concretes and mortars are virtually completely carbonated on exposure to atmospheric CO, There seems little doubt that the main chemical alteration of the surface contaminated layers of concrete at RFETS will be by carbonation
Page 25 of 57 -.. Buned Concrete at WETS-A Literatun Review
As discussed previously, carbonahon results in a significant decrease in pH from above 10 5 for buffenng by CSH and Ca(OH), to around 7-8 buffered by calcite (Figure l), this has important implications for the solubility and sorption of Pu, U and Am In general carbonated, weakly alkaline cements do not perform as well as fresh hgh pH matnces in immobilismg nuclear waste (Glasser, 1997) Carbonation of the surface of concrete results in a decrease in void space and an increase in durability Carbonation may therefore physically entrap partmlate oude contaminants, this could possibly have occurred already by atmosphenc carbonation, or may occur subsequent to bund
Controls on the redox state of groundwaters at WETS are largely undefined (EG&G, 1995), few reliable Eh measurements are available Field measurements do record dissolved oxygen, but reduced iron minerals such as sidente and pynte and carbonaceous matenals are identified in cores from the upper hydrostratigraphlc unit (EG&G, 1995), such observations are typical of the diseqwlibnum of redox reactions in natural groundwater (Stumm and Morgan, 1981) Corrosion of steel rebar exposed to groundwater in demolished concrete will act as a reducant to produce reduced groundwater conditions, the effectiveness of this reducant will depend on the groundwater flux, the rate of corrosion and the dissolved oxygen content of the inflowng groundwater It has been discussed how microbial activity can effectively catalyse redox reactions=and hydrogen produced by anaerobic corrosion could be utilised by sulphate reducing bactena, and could conceivably reduce avadable sulphate, particularly in the industnal area where 1000 mgA sulphate is measured Given the large climatic vanation at WETS seasonal vanations could conceivably result in cyclic reduction and oxidation of sulphate which gives nse to the classic sulphmc acid mediated form of mcrobial induced degradahon (Rogers et al, 1993a) Sulphide is however likely to precipitate as iron sulphide if associated wth steel corrosion and may not be available for subsequent oxidation by Throbacrllr Since sulphides are present in the local alluvium and bedrock, and there is no recorded evidence of sulphide oxidation then it would appear that microbial influences on cement degradation and controlling redox state are limited at WETS Unless some other form of reduced substrate such as organic matter is disposed wth buried concrete then it is likely that the prevailing background Eh w11 continue to apply
Modelling studies of the behaviour of large concrete structures used in near-surface low-level radioactive waste disposal repositories indicate that their gross structure wll survive intact for penods of around 1000 years (Reed et al, 1994, Gerard et al, 1997) After hstune severe crack< w11 have developed allowmg groundwater access to radioactive waste Buned concrete rubble at WETS wil not be subject to the same stresses as large intact structures, which promote the cracking and breakdown of large structures Degradation of buned concrete wll be predominantly by microscopic-scale chemical and physical processes whch w11 eventually lead to spallation of surface layers, where particulate and liquid contamination is concentrated The diffision controlled cement degradation modelling of Adenot and hchet (1997) is more applicable to the degradation of buried concrete rubble since this descnbes the chemical degradation of the concrete surface exposed to a reactive fluid Considenng that groundwater flow will be around concrete blocks rather than through the cement microporosity, diffusion will likely control the cement degradation reactions Adenot and hchet (1997) predict that for pure water dihive degradation wll extend 4 cm after 300 years in good quality concrete Carbonation, which is expected to be the dominant degradation processes for buried concrete at WETS, is likely to produce more aggressive alteration, although carbonation wll tend to reduce permeability in the concrete surface and hence reduce diffision Considenng these modelling studies and evidence from ancient cements it is likely that chemical degradation w11 extend a distance of the order of 10 cm dung the 1000 year nsk assessment penod, which will include the majonty of the surface contamination Uranium present in the aggregate is unlikely to be exposed to free-flowng groundwater during the 1000 year nsk assessment penod Page 26 of 51 @BNFLInc BudConcrete at WETS-A Literature Revlew
Spallahon of the surface layer wll increase the accessibility of groundwater to radionuclide contamination Spallahon is dependant on the nature of the degradation mechamsm, carbonation improves durability and reduces the susceptibility to freeze-thaw mechamsms Sulphate attack and accelerated leaching by microbial activity breaks down the cement matnx allowng dissagregation of the surface layer
Possible colloidal matenals generated from cement degradation are silica resulting from leachmg and carbonation of CSH and iron hydroxides produced from steel corrosion Natural colloids may exist in the background groundwater in forms such as humics or suspended clay particles denved from weathenng and erosion of bedrock claystones The additional effects of colloids generated from cement degradation should be assessed wth reference to these background colloids, and to competition for sorption of radionuclides between colloids and the concrete matnx (Bradbury and Sarrott,1994)
Leaching of Actinides in Cementitious Systems
Direct experimental measurement of the leaching and diffision of actimdes in cementitious systems is much more limited than that of more mobile radionuclides such as cesium The Swedish Nuclear Fuel and Waste Management Company (SKB) have undertaken research, from 1980 to 1990, and the results have been summansed by Albinsson et a1 (Albinsson et al, 1993) The experiments descnbed in this work involved the measurement of cesium, americium and plutonium difision into five different types of concrete The experiments were carried out over long time scales, 2 5 years for amencium and 5 years for the plutonium
The experimental technique involved takmg pre-aged concrete samples, with approximate length of 25 mm The samples were then dipped into radionuclide-spiked porewater, inside a glove box to avoid sigmficant uptake of carbonate At the end of the expenments, the concrete samples were ground, removing a 0 1 - 0 7 mm layer wth each gnnding Estimation of apparent difisivity (DJ was achieved either through activity measurements or, more accurately, through autoradiograms
The result was that no movement of plutomum or amencium could be measured (0 2 mm), despite the long time scales The D, for amencium from activity measurements was estimated at 1 - 9 x m2/s, wtule The explanation for the low difision of these actinide elements is their high sorption onto concretes The Rd for amencium sorption onto concrete varies between 1 and 10 m3/kg, while the plutonium sorption is slightly lower (R,,= 1 to 5 m3/kg) The higher sorption exhibited by amencium would suggest that its D, should be lower than plutonium, when in fact the reverse is true This may suggest that the sorption - only mechanism is not the whole story However, the fact that sorption is so high, and that the differences between the two actinides is small, means that the evidence is not clear-cut European Community sponsored research has also focused on the leaching behaviour of radionuclides from cemented waste (Vejmelka et al, 1991) Thls senes of expenments first looked at the behaviour of uraruum, plutomum, amencium and neptunium in cement samples, in a Q-bnne (MgCl, nch solution) and NaCl bnne Page 27 of 57 @BNFLInc -.. Buned Concrete at WETS-A Literature Review To accelerate the leaching process, crushed samples were used The pH of the Q-bnnekement solubon WK found to be 6 5, while the NaC1-bnnekment solution was at pH 12 5 Plutonium loadings were vaned between and 10sg/g cement, neptumum between lod and lo4 g/g cement, and uramum from lo4 to 10- ‘g/g cement These amounts roughly correspond to the levels found in real waste encapsulation The results showed, for the Q-bnne solution, a linear increase in amencium and neptunium aqueous concentrations with increasing actimde loading Thls indicates that sorption is the dominant factor in the determination of aqueous concentration The results for plutonium are not shown, but the text mentions that the same behaviour is observed for plutomum Uranium exhibits similar behaviour below loadings of 0 01 g/g cement, above thls concentration, a constant aqueous concentrabon of 5 x 10-’M is observed, which corresponds to the solubility limit of U02(0H), Thls was confirmed by the presence of a yellow precipitate In NaCl solutions, the high pH means that solubility controls the aqueous concentrations of all of the actimdes The man conclusion from th~sportion of the work is that the concentration for Am, Np and Pu is limited to le-8M to le-10 M (for both bnnes) and le-5M for urmum An important conclusion is that the behaviour of the actinides is independent of the doping technique Thus the same result is observed whether the actinide was incorporated into the cement or added later into the solution It was also discovered that the aqueous concentration of Am was not effected by the presence of cenum, indicating that cenum does not compete wth amencium for sorption sites, and that ion exchange is not a likely mechanism for radionuclide sorption onto cement matenals The kinetics of leaching from cement encapsulated matenals was also examined For Amencium, the composition of the brine had no effect on the leaching behaviour, while for plutomum, the low solubility of plutonium hydroxides meant that the expenmental concentrations were below detection limits in the NaCl bnne The results showed that the effective diffusion coefficient for Pu was 1 x m2/s, and for amencium the value was 2 x m2/s The leachability of Nd (an analogue of trivalent actimdes), tuamum, thonum and strontium have been examined in a CO, free environment (Serne et al, 1996) Crushed cement samples contaming these four radioelements were placed in deionized water, and the aqueous concentrations measured at a range of pH’s, and at different times The results showed that equilibnum was reached, in all cases, within 2 days Neodymium aqueous concentrations fell steadily with increasing pH (from 7 to 9), and were at the analytical detection limits at pH’s above 9 The solubility limits of Nd(OH), were shown to approximate the observed results well Similar results were observed for uranium, and the aqueous concentrations could be explamed by the equilibnum with CaUO, Thonum aqueous concentrations were at or below the detection limit over the whole pH range, this is probably due to the formation of the insoluble ThO,(am) phase Strontium leached concentration indicated no pH vanation, but the prospect of precipitating strontium carbonates in a real groundwater system cannot be ruled out The conclusion from this work is that solubility, rather than adsorption is the dominant factor in determining leachmg of actinides from cementitious matenals, provided the contaminants are present at > 0 15% of the cement waste form Page 28 of 57 @BNFLInc. Buried Concrete at WETS-A Literatun Review Conclusions on Leaching Experiments It is difficult to make any firm conclusions based on the limited amount of data presented here However, the followng observations can be made 1 The penetration of actimde elements into a cement waste form is limited, even when the radionuclides are initially in the aqueous phase This supports the assumption that the contamination of concretes at WETS w11 be surficial in nature, and concentrated wthin the first few millimetres of the concrete 2 The factors controlling actinide leaching fiom concrete are solubility and sorption The predominance of one over the other wll depend on the concentration of each actimde onginally present in the concrete 3 The fact that the contamination at WETS wll be surficial, rather than analogous to encapsulated waste, and leachmg determined by sorption and solubility means that sorpbon and solubility data from the literature can be used to estimate leachmg Diffusion processes are not relevant to this study Behaviour of actinides in cementitious environments The lack of much real actinide leaching data means that the chemical hehaviour of the elements in a cementitious environment must be investigated to aid elucidation of likely leaching behaviour at WETS The limited site data suggests that the contamination is likely to be particulate, particularly PuO, Thus, the first factor in determining the leach rate from the surface of the concrete is the solubility of these minerals Once in the aqueous phase, the actinides will come into contact wth the cement matnx itself, wth the possibility of sorption further retarding the movement into the groundwater Therefore, this section examines the available literature data concerned with actinide sorption and solubility in cementitious environments A hlly comprehensive review of the environmental behaviour of actinides is not reported here Rather, the essential features of actinide solubility and sorption behaviour have been explored From th~s,it is hoped, a preliminary picture of the factors influencing actinide leaching fiom concrete surfaces wll begin to emerge Summary of Conditions expected in a cementitious environment The evolution of the chemistry in cement porewaters has been descnbed earlier (Figure 1) and it has been concluded that carbonation wll have a significant effect on the behaviour of concretes close to the surface This impacts on radionuclide behaviour in two ways Firstly, formation of calcite drops the pH of the porewater solution significantly to -8, which could impact on radionuclide solubility and sorption Secondly, the formation of calcite presents a different surface to the actinides, and a consequent change in sorption behaviour Also, actinides form strong carbonate species in aqueous solution, and this can reduce sorption and increase solubility The redox conditions in the immediate environment of the contaminated concrete wll also be an important factor in the determination of actinide leaching behaviour The redox potential within the cement porewaters w11 depend on the composition of the cement Ordinary Portland cements are poorly poised, and so the redox potential is easily influenced by the groundwater in contact wth it Cement blends contsuning significant amounts of blast furnace slag will produce reducing porewaters, due to the presence of iron sulphdes, which Page 29 of 57 ~BNFLInc. m creates a SO:' /HSpoised system (as well as other sulphur couples) The result is a measured Eh of -200 mV (Glasser, 1991) However, it is unlikely that the concretes at WETS w11 contam significant quanbties of slag Corrosion of the steel reinforcements is another source of potentially low Eh values As these rebars corrode, the Eh is lowered to hydrogen liberation potential This mechanism may lower the Eh of the bulk concrete porewater, however, the majonty of the actirude contamination is on the surface of the concrete structures Therefore, it is more likely that the redox conditions of the concrete surface wll mirror very closely the redox of the groundwater According to the geochemical charactensation (EG-t;G, 1995), the redox potenbal indicates slightly oxidising conditions (90 - 320 mV) Microbial activity could potentially lower this Eh on the concrete surface, but evidence for this is lacking at present In summary, the actirudes present on the surface of concrete at WETS would, on burial, expenence alkaline conditions, due to concrete degradation However, the influence of the groundwater is likely to be large, wth carbonation of the concrete surface a very likely outcome The result would be a lower pH of -8 Redox conditions are likely to be mildly oxidising, due to the lack of redox poising from the concrete, and the consequent influence of the groundwater Speciation of Actinides General The actinide series of elements results from the filling of the 5f orbitals, the senes beginning wth thonum and ending wth lawrencium The ionisation energies of the 5f electrons play an important role in the chemical behaviour of the actimde elements The iomsation energies of the actinide elements are significantly lower than those of analogous lanthanide elements, the reason being that the 5f orbitals are considerably more shielded from the nuclear charge than the 4f electrons Thus, the 5f electrons are less firmly held than the 4f electrons in the lanthanide elements, and so more avadable for bonding The electronic structure has important implications for the available oxidation states for the actinide elements The similar energies of the 7s, 6d and 5f electrons means that multiple oxidation states are accessible, particularly for the first half of the senes In contrast, the lanthamdes are almost exclusively found in the 3+ oxidation state Table 1 (Katz et al, 1986) lists the oxidation states that uranium, plutoruum and amencium can form, with the most stable oxidation state indicated in bold Plutoruum [Rn]5f6 7s2 3,4,5,6, (7) Amencium [Rn]5P 7s2 3,43576 Table 1 Oxidation States of U. Pu and Am The oxidation state of each actinide is crucial in the determination of its mobility In general, the reduced oxidation states are less mobile, exhbiting greater sorption and lower solubility (Choppin et al, 1995) Each element will also behave differently in terms of aqueous speciation, depending on the oxidation state Page 30 of 57 @BNFL Bund Concrete at WETS-A Literature Rev& Inc Actinides in the same oxidation state have the same structure Actinide ions m the 3 and 4 oxidation states are in form of simple hydrated An3’ and An4+,although these simple iomc forms have a strong tendency for hydrolysis and polymensation unless in acidic solutions (Katz et al, 1986) Actimdes in the lugher oxidation states form oxygenated, actinyl species in solution, AnO,’ and An0,Z’ These structures are very stable in aqueous solution, possess a linear structure (Greenwood and Earnshaw, 1984) and reduce the effective charge on the central actinide ion, for example, the charge on the Pu atom has been reported as +3 2 and +2 2 for the PuOF and PuO,’ ions respectively (Silva and Nitsche, 1995) Hydrolysis is an important reaction in natural waters, and has a large influence on actinide behaviour In general, hydrolysis decreases in the order, which is expected from charge to ion size ratios Hydrolysis is important in that it can alter dominant oxidation states For example, the greater hydrolysis of An(1V) species relative to An(II1) results in greater stability for the An(1V) ion Hydrolysis is also important in terms of the forniation of polymers, (Pu colloids) wth Pu(1V) being particularly susceptible (Toth et al, 1983, Choppin et al, 1995) Plutonium polymers are very stable and not easily depolymensed, and, as the effect of, for example, concentration, temperature and ionic strength is not well understood, the erratic nature of Pu(1V) aqueous solutions can make predictions of behaviour difficult (Katz et a1 ,1986) The mobility of actinides in the environment is determined, to a large extent, by its solubility and sorption, and the results of specific expenments are presented in the followng sections The extent of both precipitation and sorption is influenced by aqueous complexation, which can reduce the extent of both processes For example, the sorption of metals onto iron oxides is reduced in CO, environment, at high pH, relative to sorption in a C0,-free atmosphere (Wate et al, 1994) In general, the trend of strengths of complexation of various ligands wth actinide ions is given by, OH-,C03’- > F , SO:., HP0:- > C1, NO,’ Thus, the main ligands that could perturb the sorption or solubility of actimdes at WETS w11 be carbonate, which is the dominant ion in WETS groundwaters, and sulphate whch is present in elevated concentrations at vmous locations (EG+G, 1995) Complexation of actinides by organic ligands is also well known (Katz et al ,1986) at WETS organic ligands could be produced by microbial activity as described earlier, or by organic contaminants Plutonium Plutonium forms a number of oxidation states which are stable in natural waters Pu(II1) is stable under acidic conditions, although it is easily oxidised to Pu(1V) PuO,’ disproportionates to Pu4’ and PuO,Z’, although there is some evidence that it may be the dominant species in solution when concentrations are low, and the chances of two PuO,’ ions interacting is very small (Choppin, 1983) Pu0,Z’ is stable but can be easily reduced, even by the action of its own a radiation (Katz et al, 1986) The dominant control of plutomum in natural waters is provided by the stability of the PuO, phase (Stenhouse, 1995), with the degree of crystallinity being the dominant factor in controlling plutonium concentrations The environmental behaviour of plutonium can be descnbed schematically, as in Figure 4 (Choppin et al, 1995) Pagc31of57 - - Buned Concrete at WETS-A Literature Review REDUCING OXIDISING 1 I Figure 4. Schematic Remesentation of the Environmental Reactions of Plutomum Uranium Uranium can form three stable oxidation states in natural waters, U“, UO,’ and UO,2’ U0,Z’ s the most stable form of uranium in aqueous solution, and is difficult to reduce (Katz et al, 1986) UO; disproportionates to U4+and UOF, whde U“ is stable wth respect to water but is slowly oxidised by mr to UO,” h et a1 (1990) note that oxygen fbgacities must be below lo4’ in order to mantarn uranium in the U(1V) oxidation state The uranyl ion is the most stable actinyl ion, wth a U-0 bond distance of 180 pm (Greenwood and Earnshaw, 1984) Uranyl forms a large number of aqueous species, of which uranyl carbonates domnate at neutral pH in natural waters Below pH 5, the uranyl ion and UO,OH+ dominate aqueous speciation The U4’, like all An4+ ions, is easily hydrolysed above a pH of around 2 9 (Katz et al 1986), and its speciation is dominated by UOH3+at low pH, and U(OH)?, and possibly U(0H); (although the presence of this species has been disputed (Rai et al, 1990) at higher pH’s Solubility control is sometimes difficult to assess due to the complex solid phase chemistry of urmum, wth over 160 uranium contaming minerals identified (Smith, 1983) Most of the uranium found in natural deposits is in the form of uraninite (UO,,, (0 O Page 32 of 57 @BNFLInc Buried Concrete at WETS-A Literature Review Americium Americium is the simplest of the actinides looked at in thls work as it forms only one oxidation state under natural conditions, Am(II1) Other oxidation states are possible but requlre either very oxidising or reducing conditions (Katz et al, 1986) Aqueous speciation is dominated by hydrolysis reactions and carbonation It has been shown (Meinrath and Kim, 1991) that, under atmosphenc conditions, hydrolysis dominates below pH 8, while at pH 8+ carbonate species are prevalent Mixed hydroxy carbonate species are also possible (Silva and Nitsche, 1995) Solubility of actinides in cementitious environments Plutonium The solubility of plutonium under cementitious repository conditions has been measured by a limited number of workers The man problem in comparisons of solubility between workers is the crystallinity of the PuO, solid phase There is evidence that crystalline PuO, is radiolytically transformed to a new hydroxide/oxide (Berner, 1995) Ewart et a1 (1992) measured Pu solubility in a Arm, atmosphere, to simulate the reducing conditions expected in a cementitious near field Figure 5 shows the resulting solubility over a pH range of 7 - 12 As can be seen the solubility of Pu is low at pH's above 8, where the aqueous concentration is below lo-'' M The higher solubility at lower pH's is due to the formation of Pu(II1) species -6 X X 7 I. X La -8 9 X X 10 x Exx X xx X 11 I 7 8 9 10 11 12 PH Figure 5 Measured Plutonium solubility under anaerobic conditions Page 33 of 57 .. - Buned Concrete at WETS-A Literature Revtcw from 8 to 10 Agam, the solubility was seen to decrease linearly wth pH below -8 (slope -0 9) Measurements of plutomum solubility under aerobic conditions are more relevant to WETS, and there are examples of these, of whch only a selection is discussed here In a study suming to deduce the effect of asphalt degradabon, plutomum solubility has been measured, at pH 12, in a concrete environment (Greenfield, et al, 1997) The resulting concentration was measured as 1 x lo-’’ M, in good agreement with the results of Ewart et al (1992) Allard and Rydberg (1983) discuss plutomum solubility in both aerobic and anaerobic waters over a range of pH’s Under aerobic conditions, plutonium concentration is controlled by the solubility of PuO,(s), wth the dominant species (in the absence of carbonate) being Pu(0H); The solubility is constant between pH 5 and 10 (as Pu(II1) species are less likely to form under aerobic conditions), wth a concentration of 109M In the leaching expements descnbed earlier (Vejmelka et al, 199l), plutomum leachability was determined to be dependent on solubility limits at pH 12 -13, and the equlibnum concentration was determined to be 1 6 x 1Oq9 M, in good agreement wth the results descnbed above It has been discussed earlier that the impact of carbonation is likely to be significant at WETS, and the effect on solubility needs to be assessed fim et a1 (1 983) observed a large increase in plutomum solubility above pH 10, in solutions contaming more than lo4 M carbonate Kim et a1 (1993) proposed that PuO, transforms into Pu(OH),C03 above pH 10, wth this phase exhibiting greater solubility The basis for thrs assumption is the drop in aqueous carbonate concentrations, indicating formation of a carbonate phase Yamaguchi et a1 (1994) also examined the effect of dissolved carbon on plutonium concentration, and they also observed increased solubility with increased carbonate levels (see Figures 6 - 7) However, Yamaguchi et a1 (1 994) could find no evidence of carbonate uptake into a solid phase, instead they proposed that the increase in solubility is due to the formation of aqueous Pu hydroxycarbonate species The mechanism is therefore in doubt, however the increased solubility is clear, and indicated that the leaching of plutomum from concrete surfaces wll be enhanced in the presence of carbonate In conclusion, the solubility of PuO, is low, even under oxidising conditions The expenmental determinations range from lo9h4 to lo-’’ M It must also be borne in mind that the crystallinity of the PuO, solid phase is a crucial factor in determmng the solubility of plutomum The expenments descnbed above used freshly precipitated or “amorphous” PuO, These wll obviously be more soluble than crystalline PuO,, and so the solubility limits could be viewed as maximum values However, the effect of a - decay can disrupt the crystal structure, reducing the crystallinity of the solid In the presence of carbonate, the solubility nses seemingly linearly above bicarbonate concentrations above 0 4 M (at pH’s below 10) and above carbonate concentrations of lo5 M, above pH 12 Page 34 of 51 Buntd Concrete at WETS-A Literature Revim 3 -4 5 -6 z X $7 A B mx 5 -9 10 11 ME03 1 WE06 1 OM-05 1 OM04 1 1 cQE-02 1 ME41 1 cQE+oo Bicarbonat. Concmtratlon, M Figure 6 Plutonium Solubility as a function ofcarbonate concentration at DH9 4.9 7 and 10 1 3 -4 -5 -6 ;-7 B A 5 X A 9 X X a. IO A A 11 0 m1 0 001 0 01 01 1 C.rbon8t. Concentration, M Figure 7 Plutonium Solubilitv as a function ofcarbonate concentration. at DH 12 and 13 Uranium The form of uranium contamination at WETS IS not known, but may be particulate or part of the concrete aggregate, and so solubility data for uranium both in terms of UO, and other minerals, has been consulted Page 35 of 57 - Buried Concrete at WETS-A Literature Review OBNFLInc - The amount of avadable data is vast, and only a fraction is presented here, although it is hoped that the man features of uranium solubility, in a WETS context, is preserved The solubility of UO, under reducing conditions has been measured by a number of workers, and the results from two studies (R~Iet al, 1990, Yajima et al, 1995) are shown in Figure 8 Both sets of experiments were camed out in inert electrolytes As can be seen, the solubility of UO, is defined by two distinct regions Below pH 4 - 5, uranium solubility increases with increasmg pH, while from pH 5+, uranium solubility is fmrly constant at M (Yajima et al) and lo-*M (R~Iet al) A slight nse in solubility can be seen at pH's above 10 This could be due to formation of the U(OH),' species or partial oxidation of U(1V) to U(V1) 1 01 0 01 + + r 0001 0' $ om1 ++ + G 0 ++ g 1E-05 ++ + 0 0 + 0 -E' lE-08 + ++ ++ + + *0 + + ++ + a 1E-07 + B *e +' 1E-08 1E49 0 *e 1E 10 2 3 4 5 6 7 8 9 10 11 2 PH Figure 8 Measured solubility of UO,- under reducing conditions Similar measurements were camed out by Ewart et al(1992), over a pH range'of 5 to 13, wth the solution composition approximated to a 9 1 Blast Furnace Slag / Ordinary Portland Cement leachate These results showed a similar constant solubility over this pH range, although the measured aqueous concentration are significantly higher than the concentrations shown in Figure 8(3 - 2 x lo-' M) There was no evidence of increasing solubility at pH lo+, in contrast to the results shown in Figure 8 The expenmental evidence, therefore, seems to point to low solubilities for uranium under reducing conditions, ranging from 2 x 10' to M, and wth no alteration of the UO, phases These expenments are useful in that they indicate the expected uranium solubility in well defined systems However, the WETS scenario calls for examination of uranium behaviour in the presence of groundwater ions, especially carbonate and under more oxidising conditions If the contamination of uranium at WETS is in the form of particulate U02, the mechmsm of dissolution under oxidising conditions can be complex U02 dissolution has been used as an analogue for the behaviour of spent fuel under repository condihons, and so its behaviour over a range of conditions has been examined by a number of workers (e g Casas et al, 1994, Torrero et al, 1994, Finch and Ewng, 1992, Wronkiewcz et Page 36 of 57 @BNFLInc. P al, 1992) A general conclusion appears to be that at least two mechanisms account for the dissolution of UO, under oxidising conditions Firstly, the surface of the UO, is oxidised, and dissolves, releasing U(V1) into solution The second, slower mechanism involves the oxidation of the bulk UO,, whch subsequently dissolves The exact mechmsm need not concern us here, but the implication is that it is the secondary minerals that control the uranium solubility under oxidising conditions Similarly, if the source of contamination is not UO,, these secondary minerals w11 be the solubility limiting phases The most apparently simple of the uranyl (U(V1)) minerals are the uranyl hydroxides, such as schoepite, which is variously given the chemical formulae UO,(OH), or UO, 2H,O It was this phase that was invoked by one of the leaching studies described earlier (Vejmelka et al, 1991) as the solubility limiting phase Schoepite exhibits the typical “U-shaped” solubility agmnst pH behaviour, wth a minimum solubility around pH 7 - 8, where the aqueous concentration equals 10” to 10“ M (Torrero et al, 1994) and lo4 to 10” ’M (Bruno and Sandino, 1989) The presence of carbonate, in even moderately low concentrations wll tend to increase this solubility, due to the formation of uranyl carbonate aqueous species (Allard and Torstenfelt, 1985) The presence of groundwater ions, such as carbonate, could result in the formation of other uranyl solid phases For example, Kat0 et a1 (1996b) found that the solubility limiting urmum phase under acidic conditions, wth a 80% - 100% CO, atmosphere, is rutherfordine OJO,CO,), while UO, was the solubility limiting phase at 0 99% CO, atmosphere The solubility of this phase, as a hction of C0:- has recently been measured (Meinrath et al, 1996), and has been shown to be crucially dependent on the carbonate concentration, as would be expected Cement leachates contain elevated concentrations of calcium, sodium, potassium and silica, and all of these species are capable of forming uranium solids Sandino and Grambow (1994) have shown that becquerelite (CaU,O,, 1 1H,O) and compreignacite (K2U6O1,1 1H,O) are formed quickly in the presence of calcium and potassium Similarly, Brownsword et a1 (1990) measured uranium solubilities in sodium and calcium hydroxide solutions, and found, above pH 7, a constant solubility of 3 x 10“ M The results did not fit the expected behaviour for a schoepite type phase, and it was postulated that sodium and calcium uranates were being formed Recently, NqU207has been identified (Yamamura et al, 1997), as has the analogous CaU,07 solid phase (Heath et al, 1997) A leachng expenment descnbed earlier (Serne et al, 1996) proposed that the observed concentrations of uranium were due to solubility limitations imposed by the CaUO, solid phase, giving aqueous concentrations of less than 10 M It must be borne in mind that no solid phase analysis was carried out, the conclusions were based solely on thermodynamic data and modelling and Berner (1 992) calls into question the stability of the CaUO, in alkaline solutions Uranium alteration by silica is a possibility in cement porewaters Finch and Ewing (1992) note that schoepite is thermodynamically unstable m waters wth even low activities of calcium and silica The mineral phase, uranophane (Ca(H,0),[(U02)(Si0,)], 3H,O) is one of the most common uranyl minerals, which may indicate that uranyl silicates are important phases controlling uranium concentrations in natural waters (Finch and Ewing, 1992) Other silicates, such as soddyite ((UO,),(SiO,) 2H,O) may also be important The formation of uranyl solid phases wthin a cement matnx have recently been examined (Moroni and Glasser, 1995), with several solubility limiting phases being identified These included becquenlite, uranophane and weeksite (K,(UO,)(Si,O,), 4H,O) along with several unidentified phases Uranium solubilities are reported to be around loe8to M, although it should be noted that the temperature in this study was 85°C More recently, a study examining the effect of silica on schoepite transformation (Sowder et al, 1996) found no evidence of Page 37 of 57 @BNFLInr - - Burwd Concrete at RFETS-A Literature Review uranyl silicates being formed, even at silica concentrations of lo3M Instead, the presence of silica was seen to retard the transformation of schoepite to becquenlite in 10-2 M and 10” M calcium systems Berner (1 992) modelled the solubility of urmum as a function of redox conditions, through consideration of urarunite, CaUO,, uranophane and “x-phase” (basically a hydrated calcium uranate) He found that under extremely reducing conditions, urarurute is the stable phase, resulting in solubilities of to 10”O M Under mildly reducing conditions, the solubility-limiting phase depends on pH, and modelled solubilities he between 10-7 to 1O-IO M The above discussion highlights the complexity of uranium in natural systems, and the uncertamty in knowledge of the identity of solubility controlling urmum solids under diffenng conditions It does appear to be evident that the presence of cement leachates does lower the solubility fiom the value expected from schoepite equilibnum M), whether through formation of calcium andor silica phases The presence of carbonate, however, appears to have the opposite effect, increasing uranium solubility, either through the formation of uranyl carbonate aqueous species or uranyl carbonate phases, such as rutherfordine Americium The solubility of amencium wll be dominated by the formation of hydroxide, carbonate and hydroxycarbonate minerals According to Choppin et a1 (1999, AmOHCO, limits the solubility of amencium when [CO:-],, > 10 l2 M (PH 6) and when [CO;-],, > M (PH 8) The number of expenments dealing wth americium solubility appear to be more limited than those for uranium, and a selection of these are presented below Atlunson et a1 (1988) measured amencium solubility in cement equilibrated waters, between pH 8 and 13 The total carbonate concentration was 3 x 10 M The results showed that the solubility was equal to M at pH’s below 10, falling gradually to lo-’’ M at pH 13 The explanation given was the transformation of americium hydroxycarbonate into amencium hydroxide, resultmg in an inflection in the solubility curve Sunilar results were observed by Ewart et a1 (1 992) also in concrete equilibrated water The solubility of amorphous Am(OH), m a carbonate free envlronment has been reported by Loida et al (1995), with the amencium concentration equalling lo4 M at pH 7, dropping to M at pH 11 The crystallinity of the Am(OH), is important, and Nitsche (1991) note that there is an order of magnitude difference between crystalline and amorphous Am(OH), (e g at pH 7, solubility is lo-’ M for the amorphous Am(OH),, compared to 1O4 for the crystalline solid) The expenments descnbed earlier (Serne et al, 1996) showed that the leaching behaviour of neodymium (an analogue for americium) was controlled by the solubility of Nd(OH),, wth a solubility of M at pH 7, and a solubility of below lom8M above pH 9 Assuming that the analogue between Nd and Am is a good one, this gives a good indication of americium leaching behaviour in a carbonate-free environment The effect of carbonation has been studied by a number of workers Meinrath and Kim (1991) consider the formation of Am,(CO,), as the result of exposing amencium to carbonated solutions The resulting solubilities range fiom 10‘ M (PH 6, carbonate lO=]M)to M (at pH 8, carbonate lO”M) In contrast, Nitsche (1992) found that, in trying to prepare Am,(CO,),, orthorhombic AmOHCO, formed in preference to Am,(c03)3 In addition, hexagonal AmOHCO, was found to form in Yucca Mountam groundwaters, at pH 5 9 (Nitsche, Page 38 of 57 ~BNFLInc Buned Concrete at RFETS-A Literature Rcvlew 1991) (the orthorhombic solid formed at pH 7 and 8 4) The solubilihes in these Yucca Mountam groundwaters was determined to be -1 O4 M, wth solubility nsing slightly as pH is increased Other determinations of americium solubility includes Allard and Torstenfeldt (1 989, where the calculated aqueous concentration was determmed to be M in marl groundwater, and Vejmelka (1991), where the leachable concentration of amencium was equal to the solubility limit at pH 12 - 13 (2 x lo-’’M) In conclusion, amencium solubility is dependent on both pH and carbonate There is evidence of phase changes from amencium carbonates, or hydroxycarbonates below pH 9, to Am(OH), at hgher pH Solubilibes appear to vary from around 10” M at neutral pH’s to 10 lo M at alkaline pH’s Sorption of actinides onto cements Sorption data is normally presented in terms of dzstrzbutzon ratios, or &’s (or &’s) The distnbution ratio can be easily extracted from batch sorption expenments, through use of the followng expression, Quantity of radionuclide sorbed per unit mass of solid R, = Equilibnum concentration of radionuclide in solution =(“‘;2c2)’ M where C, - initial aqueous concentration of radionuclide - c* final aqueous concentration of radionuclide V - volume of solution M= mass of solid phase Cementitious matenals are likely to be good sorption substrates for elements whose dominant sorption mechanism is surface complexation The specific surface area of cement matenals are high, ranging from 55 to 200 m2/g (Lea, 1980, Bradbury and Sarott, 1994), suggesting that sorption capacity is hgh As has been descnbed above, the actimde elements have a significant tendency to hydrolyse, wth high first hydrolysis constants (IC,,), and high charge to bond length (Ad) ratio (Baes and Mesmer, 1986) There is a strong correlation between distnbution ratios on cement, and K,, and z/d values (Bradbury and Sarott, 1994) It would therefore be expected that uranium, plutonium and amencium would exhibit hgh distnbution ratios The effect of cement composition needs to be addressed Several studies (Allard et al, 1984, Allard, 1985, Atkinson et al, 1988) have suggested that the specific composition of the cement has little or no effect on the sorption of readily hydrolysed elements, such as actinides There is evidence that the sorption of non- hydrolysed species such as Ra, Cs and I is crucially dependent on the cement composition (Heath et al, 1996, Holland and Lee, 1992, Glasser et al, 1997) In particular, the sorption of Cs (whlch is poorly sorbed by CSH phases) is claimed to be extremely sensitive to the composition of the ballast (aggregate) (Allard, 1985) or aluminium substitution into the cement (Glasser et al, 1997) From the evidence, it appears that a good first approximation is that the specific cement composition has little impact on the sorption of actinides However, it must be borne in mind that most, if not all, of the experimental determinations of sorption onto cements were cmed out at pH 12 and above At this pH, and under the low Carbonate levels conszdered, actinides wll be hydrolysed and sorption is likely to be close to Page 39 of 57 ~BNFL - - Buned Concrete at WETS-A Literature Rcvlcw Inc W 100% Under these conditions, vanations in sorption due to the composibon of the cement could be masked Ifthe porewater close to the surface ofthe concrete is influenced by the groundwater composition, pH wll be lower than 12 and therefore, sorption may be lower and compositional vanation more important Distnbution ratios gleaned from the literature do indeed show lugh distnbution ratios for actinides on cement I& Below is listed a summary of some of the relevant values for uranium, plutomum and amencium onto cements oxidising conditions 8 U(1V) sorption value Heath et al, 1996 1996 62 UWI).I somtion 1) value I Heath et al, 01-2 Recommended values for oxidising I Bradbur; and Sarott, 1994 conditions 1 1-5 Recommended values for reducing Bradbury ahd Sarott, 1994 conditions Plutonium 1 2 - 12 6 Sorpbon onto seven cement blends, Allard et al, 1984 oxidising- conditions 10 - 30 Vmation due to vanation in solid- Atlunson et al, 1988 liquid ratio 66 1996 PU(IV).I Heath et al, J 1-5 Recommended values for both Bradbury and Sarott, 1994 oxidising and reducing conditions Amencium 25-25 1 Sorption onto seven cement blends, Allard et al, 1984 oxidising conditions 7- 60 Vanation due to examination of Atkinson et al, 1988 I-Darticle size I 11-5 I Recommended values for both I Bradbury and Sarott, 1994 I I I oxidising and reducing conditions I . Table 2 Summary of distnbution ratios for actimdes onto cements Thus, the actinides, particularly amencium, apparently exhibit extremely strong sorption on cementitious materials There is no apparent effect of pH, although the expenments were all cmed out above pH values of -12 5 In explaining the sorption behaviour observed in these experiments, all of the authors have invoked surface complexation onto CSH phases as the most likely sorption mechanism This is likely to be true for cements that have not been extensively leached or altered by interaction with groundwaters Upon hydration, CSH forms on the outside of the cement powder particles, on the boundary between the interstitial water and the solid phases (Bradbury and Sarott, 1994) The CSH phase is almost amorphous and contains most of the microporosity ofthe cement Accordingly, “much ofthe sorption potential associated wth cement mses from the CSH micropore network and its concomitant high surface area” (Glasser, 1992) Page 40 of 57 @BNR Bund Concrctc at WETS-A Literature Review Inc. ? However, the situation is likely to be different for a cement that has been leached, or more relevant to WETS, carbonated Carbonation results in the formation of calcium carbonate, whch tends to fill space (and block cracks), and form a relatively dense skm of solids, effectively isolating the cement phases from the aqueous phase (Glasser et al, 1997) This means that calcium carbonate solids wll become the dominant sorption substrate, wth a resultant drop 111 sorption For example, Stenhouse (1995) reports & values of 0 02 m3/kg, 0 4 - 5 m3kg and 5 m3kgfor uranium, plutonium and amencium sorption on calcite These distnbution ratios are sigmficantly lower than corresponding cement &'s, particularly for uranium The precipitation of calcium carbonate minerals at the cement surface may induce the formation of co- precipitates wth the actimde elements already present co-precipitation of trace elements within a calcium carbonate solid phase is a well-investigated phenomenon (e g Comans and Middelburg, 1987), wlth strontium predicted to be particularly effected (Plummer and Busenberg, 1987) An unportant consequence of co- precipitahon is that, in contrast to surface sorptxon, the contaminant is held irreversibly, and so is inaccessible to the aqueous phase This obviously has important consequences for the leachmg of radionuclides from the surface of buried concrete, and therefore needs to be assessed Carllsson and Aalto (1996) have recently examined the co-precipitation of uranium with calcium carbonate The results indicated that, under the expenmental conditions (PH 9 6 and 8 5, in both mtrogen and CO:, atmosphere) urmum did not co-precipitate wth calcium carbonate It was postulated that dominant species in solution, uranyl carbonates, were not of suitable size and charge to be incorporated into the calcite structure This result confirmed earlier work by Carroll and Bruno (1991) which showed that co-precipitation did not occur over the range of experimental conditions In contrast to these experimental results, recent modelling work (Curti, 1998) predicts that the tnvalent and tetravalent actinides will be significantly affected by co-precipitation wth secondary calcium carbonate minerals The implication to the behaviour of Pu(1V) and Am at WETS is clear However, this work must be viewed with caution The results presented are purely the result of modelling, based on a simple distnbution law model This in itself is not reason to be cautious However, the important parameter, the partition coefficient, has been estimated from consideration of the solubility products of pure metal carbonates, & values and the ionic radii of the metals It is not clear at this point as to the validity of th~sapproach, and expenmental confirmation is required The study does, however, emphasise the potential effect of calcite co- precipitation, and could be an important mechanism in the determination of leaching behaviour at WETS The conclusion from this survey of sorption processes is that actinides are strongly sorbed onto cement phases, wth americium particularly strongly sorbed It would therefore seem that if uranium, plutonium and amencium were sorbed onto the surface of concrete the concentration released into the aqueous phase w11 be limited However, the surface of a burred concrete structure will be extremely susceptible to groundwater influences, particularly carbonation Carbonation produces precipitates of calcium carbonate, for which the actinides have a lower affinity, resulting in less retention Co-precipitation, which would effectively retard the release of contamination irreversibly, does not appear to be a viable mechamsm for uranium, and, although a theoretical study has demonstrated the possible consequences for plutonium and americium, firher experimental work is needed before firm conclusions can be made Page 41 of 57 OBNFLInc -.. BWed Concttte at WETS-A L~tcratuttRCVICW Summary and Conclusions A literature search and review has been undertaken regarding the processes controlling the behaviour of plutonium, uranium and amencium in contaminated concrete at WETS The review has concentrated on two aspects degradation of the concrete surface layer, where the bulk of the contaminahon exists, and a review of expertmental studies of leachmg of plutomum, urmum and amencium from cement-based matenals It was qwckly apparent that there was a sigmficant lack of data on the latter Consequently literature on radionuclide solubility and sorption have been examined to provide a background understanding of the behaviour of actinides in cementitious environments Concrete Degradation A review has been cmed out on literature concerning the long-term degradationcofcement and concrete used in radioactive waste encapsulation, and in repository structural designs, together wth more general literature hs on the surface-dominated degradation of concrete and cement The findings of review have been considered together wth site specific geochemical data to understand the likely behaviour of the surface layers of contaminated concrete, where the majority of the contamination exists The man form of concrete degradation occmng in WETS concrete is likely to be carbonation of the surface During hsprocess carbon dioxide present in am, soil gas and dissolved in groundwater reacts wth calcium phases (Ca(OH), and CSH) wthn the cement matnx to produce calcium carbonate and silica Supporting evidence for this process comes from expenmental studies, and examination of ancient analogues of modern OPC concrete From modelling studies and examination of long-term carbonation it is likely that carbonation of the concrete surface wll extend to a depth of around 10 cm in the 1,000 year perrod considered and wll thus include the zone of contamination of plutomum, uranium and amencium resulting from rutrate solution and particulates Since the background groundwater at WETS is dilute other forms of concretedegradation such as sulphate and chlonde attack are unlikely to be effective In the industnal area of the site sigruficant sulphate concentrations are measured (1000 mg/l) at which the ettringite form of sul$ate promoted attack might be expected if these sulphate concentrations were maintained in the vicinity of the disposed concrete Other degradation processes to consider are the corrosion and expansion of steel rebars, which wll produce cracking of large blocks of concrete, corrosion wll occur at maximum rate since corrosion protection offered by the high pH of unleached concrete wll not be effective in close contact with groundwater Microbial induced degradation of concrete has been considered, particularly since the contamination is located at the concrete surface where microbial growth is most likely There is however no clear evidence that microbial activity will be significant, although substrates for microbial growth are present at WETS (e g pyrite, sulphate) Degradation by carbonation is unlikely to disrupt the mechanical structure of the surface layers of concrete since the calcium carbonate formed is an effective cement whch has been shown to survive for 2,000 years in ancient structures Porosity is also reduced by carbonation whch wll reduce the effect of purely physical Page 42 of 57 @BNFLInc. Buned Concrete at WETS-A Literature Review degradation processes such as freeze-thaw Freeze-thaw and mechanical erosion of the concrete surface wll be effective given the seasonal climatic vanahon and the topography at WETS, such factors wll also depend on the nature of disposal such as depth ofbmal Gross breakdown of concrete blocks is most unlikely over a penod of 1000 years and thus access to urmum contamination in aggregate by groundwater wll be restrrcted Growth of carbonate could conceivably entrap particulate contamination in the surface layer and restrrct access by groundwater Degradation by sulphate attack or microbial action is more likely to disaggregate the surface layers of concrete and is likely to provide better access of groundwater to contamination Plutomum, uramum and amencium contamination could be mobilised from the surface of concrete by colloidal matenal generated from concrete degradation The most likely colloids formed from degradation of WETS concrete are silica colloids resulting from carbonation of CSH phases and iron hydroxide colloids formed from steel corrosion Studies of colloid formation in cement leachate has been examined in carbonate free expenments, however these are not applicable to WETS as the CSH colloids produced are unlikely to be stable in the presence of carbon dioxide The ability of colloids to significantly increase radionuclide mobility in cementitious system has been brought into question recently Chemical degradation of cement and concrete is critical to the mobility of plutonium, uranium and amencium In addition to changes in cement mineralogy chemical degradation controls the composition of the local flwd which influences actinide solubility and sorption pH redox potential (Eh) and carbonate content are all important factors controlling solubility and sorption The most significant effect of concrete degradation is a reduction in pH from over 12 in fresh cement pore fluid containing free alkalis to pH-1 0 5 buffered by CSH phases and finally to a pH -8 buffered by carbonate phases The aqueous carbonate content is controlled by CO, partial pressure which is influenced by external controls such as plant and microbial respiration in the soil zone and exchange with the atmosphere Cement and concrete will also regulate the CO, partial pressure in a closed system Degradation ofthe cement matrix is unlikely to influence Eh, however corrosion of steel rebars is likely to produce anaerobic conditions Overall, from information in the literature it is possible to qualitatively predict the degradation behaviour of concrete at WETS to provide background information for evaluating the chemical controls on leaching, solubility and sorption of plutomum, urmum and amencium Some areas of uncertamty that have amen dmng this review are Consideration of the effect of sulphate contamination in groundwater in the industrial area, is it representative of long-term composition7 The potential role of microbiological processes in influencing concrete degradation and in controlling redox conditions The potential for generation of actinide sorbing colloids during carbonation ofconcrete The lack of understanding of processes of redox control and colloid transport are uncertamties which are not necessarily confined to this study but are significant to radionuclide mobility in general at WETS Leaching, solubility and sorption of plutonium, uranium and americium in cementitious systems Expenmental measurements of the leaching of plutonium, uranium and amenciwn from cement matrices are very limited in extent The mam reason appears to be the low mobility of actinides in cementitious Page 43 of57 ~BNFLInc environments, wth a consequent need for long bme scale expenments Literature on radionuclide solubility and sorption have been consulted, to provide an understanding of the likely behaviour of actimdes in a cemenbtious environment From the results of hsliterature review, the followng conclusions can be made Firstly, the penetration of ttramum, ammcium and plutomum into intact concrete is very low, even when the radionuclides are in aqueous solution This confirms the supposibon that any concrete contamination at WETS wll be surficial In addition, the low diffisivity means that any uranium present in the concrete aggregate should not leach into the groundwater, in significant quantities, over a 1,000 year time scale The diffusion coefficients presented by two sets of workers are reasonably consistent, especially so when the low mobility and thus difficulty in measurement is taken into account The diffusion coefficient for plutonium was determined to be 2 x 10 l7 m2/s (Albinsson et a1 1993) and 1 x m2/s (Vejmelka et al, 1991) For amencium, the corresponding D,’s are 0 3 - 1 8 x m2/s (Albinsson et al, 1993) and 2 x lo-’’ m2/s Vejmelka et al, 1991) The mechanisms controlling radionuclide leaching from cement or concrete m sorption onto the cement matnx and precipitation of solid phases The dominant mechanism w11 depend on the imtial concentration of the radionuclide In general, radionuclides below the solubility limit wll exhibit a linear increase in leached concentration, as sorption dominates (assuming linear sorption) Above a certam concentration, solubility limits wll determine the leached concentrations, this value w11 correspond to the maximum extent of leaching, and will be fixed, unless geochemical conditions change The immediate environment of the concrete wll be charactensed by high pH, wth the leachate contzuning elevated concentrations of calcium, silica and, imtially at least, alkalis, such as Na and K In addition, carbonation is likely to occur, inducing calcite precipitation All of these factors, as well as the composiaon of the groundwater are likely to influence the leaching behaviour of the actimdes The solubility of plutonium, uranium and amencium, over a range of conditions, have been examined In general, solubility is expected to be low at high pH’s and low carbonate Plutonium, for example, is likely to have a solubility between M and lo-’’M The effect of carbonation is llkely to be crucial, wth plutomum exhibiting increased solubility when bicarbonate concentrations are above lo4 M Urmum and amencium are likely to change phases as a result of carbonation The effect of calcium, silica and potassium on uranium solubility has been examined, with the conclusion that formation of phases such as uranophane wll tend to reduce solubility The formation of these phases under site specific circumstances, however, w11 have to demonstrated There is lack of data on similar phases of amencium and plutonium, should they exist Sorption of actinides onto cements is likely to be high, as would be expected by easily hydrolysed elements, and I& values are as high as 30 m3/kg All of the measured ‘s have, however, been measured at very high pH’s (12 - 13), which is more akin to the environment of cement - encapsulated waste It is not clear whether these conditions will apply to the immediate environment of surface contaminated actinides In particular, carbonation w11 result in the precipitation of calcite, which wll present a new sorption substrate to the radionuclides Sorption onto calcite appears to be less strong than sorption onto cement phases The possibility of irreversible, co-precipitation of the actinides wth calcite has to be considered The expenmend data suggests that co-precipitation of uranium (VI) with calcite does not occur, probably due to the size and charge of the uranyl carbonate aqueous species However, it has been shown, theoretically, that co- precipitation could have a major impact on radionuclide migration Page 44 of 57 @BNFL Buned Concrete at WETS-A Literaturr Review Inc From this literature review, it is clear that there are a number of dominant factors that wll influence the leachmg of radionuclide - contaminated concrete A preliminary list of these factors is listed below, hs I Nature of contamination - w11 determine the penetration of the radionuclide into the concrete (1 e aqueous contamination wll penetrate more deeply), and the initial solid phase present I1 The amount of contamination - different mecharusms (sorption or solubility) w11 operate at different concentrations If the contamination is predominantly sdicial, it is more difficult to define a g/g concentration I11 Geochemical Conditions - the conditions in the immediate environment of the contamination is crucial in the determination of leach behaviour e g A interaction of the groundwater wth the concrete and concrete porewaters B pHandpe the mobility of radionuclides in any environment is crucially dependent on these two parameters C ligands- e g carbonate, calcium, alkalis, sulphate, orgarucs D alteration of cement minerals - co-precipitation E nature of solubility limiting phase, if any In summary, it is likely that the leaching behaviour of uranium, plutonium and amencium, on purely chemical considerations, is controlled, ultimately, by the equilibnum wth a solubility controlling solid phase The exact nature of this solid phase, and the value for its solubility limit w11 be complex, and wll depend on all the factors mentioned above Page 45 of 57 . - Bund Concrete at RFETS-A Literature Revlm Bibliography Adenot, F. and Bml, M. (1992) Modelling ofthe corrosion ofthe cement paste by deionized water Cement and concrete research, 22 (2-3) p489-496 Adenot, F. and Rwhet, C. (1997) Modelling ofthe chemcal degradation of a cement paste In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L ,Young, J F ,Eds E & FN Spon, London p34 1-349 Akers, D.W., Kraft. N.C. and Mandler, J.W. (1994) Compression and immersion tests and leaching of radionuclides, stable metals, and chelatmg agents from cement-solidified decontammauon waste collected from nuclear power stations Nuclear Regulatory Commission NUREG/CR-620 1,225~ Albinson, Y., Andersson, K., Barjesson, S. and Allard, B. (1993) Dimion of radionuclides in concrete/bentonite systems, SKB Techcal Report 93-29 Alcorn, S R., Coons, W.E., and Gardiner, M.A. (1990) Estimation oflongevity'of Portland cement grout using chemical modeling techniques Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management XI. 176, p 165 - 174 Allard B., Elliason L., Hoglund S. and K Anderson (1984) Sorption of Cs, I and actirudes in concrete systems, SKBKBS Technical Report SKBKBS - TR 84-15 Allard, B (1985) Radionuclide Sorption on Concrete NAGRA Technical Report NTB 85-2 1, Allard, B and Rydberg, J. (1983) Behaviour of Plutonium in Natural Waters in Plutonium Chemistry (ed W T Carnal1 and G R Choppin) ACS Symposium Senes Volume 216 Allard, B. And Torstenfelt, B. (1985) Actinide Solubilities and Speciation in a Repository Environment Nagra Technical Report 85 - 18 Amaya, T., and Suzulu, K. (1992) New Techque to evaluate the surface degradation ofcementanaceous matrix Proceedings of internabonal hgh level radioactive waste management (IHLRWM) conference, Las Vegas, American Nuclear Society publishers J. Appelo, C.A. and Postma, D. (1993) Geochemistry,groundwater andpollution Balkema 536p Atkins, M. Bennett, D.G., Dawes, A.C., Glasser, F.P., Kindness, A. and Read, D. (1992a) A thermodynamic model for blended cements Cement and concrete research, 22 (2-3) p497-502 Atkins, M., Bennet, D., Dawes, A.and Glasser, F. P. (1992b) A Thermodynamic Model for Blended Cements DOE(UK) Report DoE/HMIP/RR/92/005 Atkins, M., Glasser, F.P. and Kindness, A. (1992~)Cement hydrate phases solubility at 25OC Cement and concrete research, 22 (2-3) p24 1-246 Atluns, M.,Damidot, H. and Glasser, F.P., (1994) Performance ofcementitious systems in the repository Materials Research Society Symposium, ScientiJc Basis for Nuclear Waste Management WII 333, p 3 15 - 326 J. Athnson A., Ewart, F. T., Pugh, S. Y. R., Rees, H., Sharland, S. M., Tasker, P. W.And Willuns, J. D., (1988) Experimental and modelling studies of the near-field chemistry for Nirex repository concepts UK Nirex Ltd Report NSS/R104 Page 46 of 57 @BNFLInc Buried Concrete at WETS-A Literature Rcvlew Atlunson, A. and Hearne, J.A. (1990) Mecharustic model for the durability of concrete bamers exposed to sulphate-bearmg groundwaters Materials Research Society Symposium, Scientific Basis for Nuclear Waste ManagementXIII 176, p149-156 Baldwin, C.E., and Navratil, J.D. (1983) Plutonium process chemistry at rocky flats In Carnall, W T , Choppin, G R (Eds) Plutomum Chemsitry ACS Symposium Series 2 16 Amencan Chemical Society, Washington p 369-379 Bateman, K., Coombs, P., Noy, D.J., Pearce, J.M. and Wetteon, P.D. (1998) Numerical modeling and column expenments to simulate the alkaline disturbed zone around a cementitious radioactive waste repository Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management XYI 506, p 605-61 1 Bennett, D.G., Read, D., Atluns, M. and Glasser, F.P. (1992) A thermodynarmc model for blended cements I1 Cement hydrate phases, thermodynamic values and modelling studies Journal of Nuclear Materials, 190 p 3 15-325 Berner, U.R. (1 987) Modelling porewater chemistry in hydrtaed Portland Cement Materials Research Society Symposium, Scientijic Basis for Nuclear Waste Management X 84, p3 19-330 Berner, U.R. (1992) Thermodynamic modelling ofcement degradation impact ofredox conditions on radionuclide release Chemistry ofcements for nuclear applications Cement and concrete research, 22 (2-3) p465-475 Berner, U (1995) Estimation of Solubility Limits for Safety Relevant Radionuclides NAGRA Technical Report 94-08 Borjesson, K.S., Emren, A.T. (1993) SOLISOL - A program using PHREEQE to solve solid solutiodaqueous equilibriums Computers and Geosciences 19 pl065- 1070 Bradbury, M.H. and Sarott, F.-A. (1994) Sorption Databases for the Cementitious Near-Field of a L/ILW Repository for Performance Assessment NAGRA Technical Report NTB 93-08 1 17p Brierley, C. L. and Brierley, J. A. (1997) Microbiology ofthe metal mining industry Manual of Environmental Microbiology, pp 830-841 eds Hurst et al, ASM Press, Washmgton, USA Brownsword, B., Buchan, A. B., Ewart, F. T., McCrohon, R., Ormerod, G. J., Smith-Briggs, J. L. and Thomason, H. P. (1990) The Solubility and Sorption of Uranium (VI) in a Cementitious Repository Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management XIII 176 p577 Bruno, J. and Sandmo, A. (1989) The Solubility of Amorphous and Crystalline Schoepite in Neutral to Alkaline Aqueous Solutions Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management XII 127 p87 1 Carde, C., Francoise, R. and Ollmier, J-P. (1997) Microstructural changes and mechanical effects due to the leaching of calcium hydroxide fiom cement paste In Mechanisms Chemical Degradation Cement-Based J of of Systems Scnvener, K L , Young, F ,Eds E &FN Spon, London p30-37 An Carlsson, T. and Aaalto, H. (1996) Co-precipitation of Sr, Ni and U wth CaCO, expenmental study Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management XIX 412, p 647 - 65 1 Carroll, S. A. (1992) Mechanisms controlling NdOHC0,-CaCO, solid solution formation Water-Rock Interaction, Kharaka and Meast (eds) Balkema, Rotterdam p89-92 Page 47 of 57 It f Y c* JIbU ~BNFL - - Buned Concrete at WETS-A Literature Review Inc P Carroll, S. A. and Bruno, J (1991) Mineral - Solution Interactions rn the U(V1) - CO, - H,O System Radiochimica Acta, 52/53, pl87 Cary, J.W. Gee, G.W. and Whyatt, G.A. (1991) Waste storgae in the vadose zone affected by water vapor comdensation and leachmg Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management XIV, 212 p 871-878 Casas, I., GimCnez, J., Marti, V., Torrero, M. E. and de Pablo, J. (1994) finetic Studies of Umrradiated UO, Dissolution under Oxidising Conditions in Batch and Flow Expenments Radiochimica Acta 66/67 p23 Chambers, A.V. (1995) Modelling radionuclide transport in cracks through cemented radioactive waste Materials Research Society Symposium, Scientific Basis for Nuclear Waste Management XVIII 353, p89 1- 898 Choppin, G. R. (1983) Aspects of Plutonium Solution Chemistry in Plutonium Chemistry (ed W T Carnal1 and G R Choppin) ACS Symposium Senes Volume 2 16 Choppin, G., Rydberg, J. and Liljenzin, J. 0. (1995) Radiochemistry and Nuclear Chemistry 2nd Edition Butterworth-Heinemann (publishers) Cohen, M.D. (1987) Damage mechanism of cyclic freezing-thawng in sulfur concrete Cement and concrete research, 17 (2), p 357-360 Comans, R N. J., and Middelburg, J. J., (1987) Sorption of trace elements on calcite Applicability of the Surface Precipitation Model Geochimica et Cosmochimica Acta, 51, p2587 Constantinou, A.G. and Scrivener, K.L. (1997) Microstructural examination of the development of corrosion in reinforced concrete In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L , Young, J F , Eds E &FN Spon, London p134-142 Crammond, N.J. and Halliwell, M.A. (1997) Assessment of the conditions required for the thaumasite form of sulphate attack In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L , Young, J F ,Eds E &FN Spon, London p193-200 Croney ST (1985) Leachability of Radionuclides from Cement Solidified Waste Forms Produced at Operating Nuclear Power Plants Nuclear Regulatory Commission, Washngton, DC Office of Nuclear Regulatory Research NTIS/NUREG/CR-4 18 1,128~ Curti, E. (1998) Coprecipitation of radionuclides dmng cement degradation A Preliminary modelling study for the Swiss LIILW repository Materials Research Society Symposium, Scientific Basis for Nuclear Waste Management XXI 506, p3 13-320 Damidot, D., Atkms, M., Kindness, A. and Glasser, F.P. (1992) Sulphate attack on concrete limits of AFt stability domain Cement and concrete research, 22 (2-3), 229-234 Dearlove, J.P.L., Green, D.C. and Ivanovich, M. (1989) Uranium transport and the partitioning of U, Th and Ra isotopes between solid and aqueous phases in the Krunkelbach mine, Federal Rebulic of Germany Materials Research Society Symposium, Scientrfc Basis for Nuclear Waste Management XII 127, p927-932 Delagrave, A., Gerard, B. and Marchand, J. (1997) Modelling the calcium leaching mechanisms in hydrated cement pastes In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L , Young, J F , Eds E &FN Spon, London p38-49 Dicke, C.A., and Smith, RW. (1993) Mobility of chelated radionuclides in engineered concrete bamers Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management XVI 294, p 279-284 Page 48 of 57 @BNFL Buncd Conmte at WETS-A Literatun Review Inc DOE (1996) Final Rocky Flats Cleanup Agreement Umted States Environmental Protection Agency Region VI11 and The Sate ofColorado 84p DOE (1995) Contaminated Concrete Occurrence and emerging technologies for DOE decontamination, U S Department ofEnergy, Office of Environmental Restoration and Waste Management, Office ofTechnology Development, Report DOE/OR0/2034 Duerden, S., Glasser, F.P., Goldthorpe, K., Pedersen, J., Quillan, K., Ross, D., Stronach, S.A. and Tyrer, M. (1997) Chemistry and Performance of blended cements and backfills for use in radioactive waste disposal Materials Research Society Symposium, Scient fic Basis for Nuclear Waste Management XX 465, p287-294 EG &G (1995) Groundwater Geochemistry Report for the Rocky Flats Environmental Technoology Site Volume I11 ofthe Sitewde Geoscience Characenzation Study EG&G Rocky Flats Inc, Golden Colorado Engkvist, I., Abinsson, Y, and Engkvist, W.J. (1996) The long-term stability of cement Leachmg tests SKB Technical Report 96-09 39p Ervin, P (1998) Personal Communication, Telephone conference 19th May, 1998 Ewart, F. T., Smith-Briggs, J. L., Thomason, H. P. and Williams, S. J. (1992) The Solubility ofActinides in a Cementitious Near - Field Environment Waste Management 12, p24 1 Faucon, P., Jacquinot, J.F., Adenot, F., Virlet ,J., and Petit, J.C. (1997) Influence of the Degraded Surface Layer on the Long Term Behaviour of Cement Pastes Materials Research Society Symposium, ScientlJic Basis for Nuclear Waste Management XX 465, p295 - 302 Fayek, M., Kyser, T. K., Ewing, R C. and Miller, M. L. (1997) Uraninite - Water Ineractions in an Oxidising Environment Materials Research Society Symposium, Scientijic Basis for Nuclear Waste Management XY 465 p465 Ferraris, C.F., Clifton, J.R., Stutzman, and Garboczi, E.J. (1997a) Mechanisms ofdegradation ofportland cement-based systems by sulfate attack In Mechanisms of Chemical Degradation of Cement-Based Systems J Scnvener, K L ,Young, F ,Eds E &FN Spon, London p185-192 Ferraris, C.F., Clifton, J.R., Garboczi, E.J. and Davis, F.L. (1997b) Stress due to alkali-silica reactions in mortars In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L ,Young, J F , Eds E &FN Spon, London p75-82 Finch, R. J., and Ewing, R. C. (1992) The Corrosion of Uraninite under Oxidising Conditions Journal of Nuclear Materials 190, p133 Franz, E-M., Fuhrmann, M. and Bowerman, B. (1994) Proposed waste form performance cntena and testing methods for low-level mixed waste USDOE Report DOEMWIP-30, 165p Fujiwara Y; Maruya T; Owaki E (1992) Degradation of concrete buried in soil with saline ground water Nuclear engineering and design, 138 (2) p 143-150 Fujiwara, A,, Tashiro, S., Takemura, T. Sakogaichi, K., Yokomoto, S. and Katsuyama, K. (1995) Macro and micro observations on mortar alteration mechanisms under various solution conditions Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management WIII 353, p899-906 Gardiner, M.P., Holtom, G.J. and Swanton, S.W. (1997) Influence of colloids, microbes and other pertubations on the near-field source term Presented at the RSC/Nirex Loughborough Symposium, Loughborough, UK, December, 1997 (to be published as a Nirex Science Report) Page 49 of 57 @BNFL . Buried Concntc at WETS-A Literature RCVICW - Inc Gerard, B. Didry, O., Marchand, J., Breysse, D. and Hornain, H. (1997) Modelling th elong-term durability of concrete for radioactive waste disposals In Mechanisms of Chemical Degradation of Cement- Based Systems Scnvener, K L , Young, J F Eds E &FN Spon, London p33 1-340 Glasser, F.P. (1992) Progress in the immobilizabon of radioactive wastes in cement Cement and concrete research, 22 (2-3) 201-216 Glasser, F.P. (1997) Chemical attack on cements in nuclear repositones In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L ,Young, J F ,Eds E &FN Spon, London p323-330 Glasser, F.P., Adenot, F., Harris, Amarantos, S.G., Noe, M., Gonzalez de la Hubra, M. and Van Velzen, L. (1997) Charactensation and performance of cemetitious products in radioactive waste repositones Proceedings of 4'h European Conference on management and disposal of radioactive waste EUR 17543 EN ~179-194 Glasser, F.P., Atluns, M., Kindness, A., Pointer, C., Cowie, J.,Jappy, T.,Macphee, D. Beckley, N., Carson, S., Wilding, C.R., Mattingley, N.J., Evans, P.A., and McHugh, G. (1990) Charactensation and modelling of blended cements and their application to radioactive waste immobilization U K Department of Environment Report, DoE/HMIP/RR/90/062 Greenfield, B. F., Ito, M., Spindler, M. W., Williams, S. J. and Yui, M. (1997) The Effects of the Chemical and Radiolytic Degradation of Asphalt on Plutomum Solubility Materials Research Soczety Symposium, Scientific Basis for Nuclear Waste Management XX 465 p72 1 Greenwood, N. N., and Earnshaw, A. (1985) Chemzstry of the Elements Pergamon Press (publishers), 1542p Hall, G. R. (1989) Control of Microbially induced corrosion of concrete in wastewater collection systems, Material Performance, 28(10) 45-49 Hamilton, M.A., Rogers,R.D., Veeh, R. and Zolynslu, M. (1996) Evaluation of microbially influenced degradation of massive concrete structures Materials Research Society Symposium, Sczentfic Basis for Nuclear Waste Management XM 412, p 469 - 474 Heath, T. G., Pilkington, N. J., Tweed, C. J. and Williams, S. J. (1997) Radionuclide Solubility at High pH Presented at the RSCMirex Loughborough Symposium, Loughborough, UK, December, 1997 (to be published) Heath, T.G., Ilett, D.J. and Tweed, C.J. (1996) Thermodynamic Modelling of the sorption of radioelements onto cementitious materials Muterials Research Society Symposzum, Scientific Basis for Nuclear Waste Management XM 412, p443 - 449 Hoglund, L.O. (1992) Some notes on ettrrngite formation in cementitious matenals Influence of hydration and thermodynamic constraints for durability Cement and concrete research, 22 (2-3) p 2 17-228 Holland, T. R. and Lee, D. J. (1992) Radionuclide Getters in Cement Cement and Concrete Research, 22 p247 Houst, Y.F. (1997) Microstructural changes of hydrated cement paste due to carbonation In Mechanisms of Chemical Degradation of Cement-Based Systems Scrivener, K L Young, J F ,E& E &FN Spon, London p90-97 Page 50 of 57 Buned Concrete at WETS-A Literature Review Jakubick, A.T. Gillham, R.W., Kahl, I. and Robin. M. (1987) Attentuation of Pu, Am, Cs and Sr mobility in concrete Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management X 84, ~355-368 Jiang, W. and Roy, D.M., (1994) Ancient analogues concerning stability and durability of cementious waste form Materials Research Society Symposium, ScientiJic Basis for Nuclear Waste Management xyll 333, p 335 - 340 Kato, S., Yanase, Y. and Honda, T. (1996a) Leachng and adsorption charactenztics of radionuclides in activated concrete waste Proceedings of IRPA9 1996 international congress on radiation protection p 354- 356 Kato, Y., Kimuar, T., Yoshida, Z. and Nitani, N. (1996b) Solid - Liquid Phase Equilibria of Np(V1) and U(V1) under Controlled CO, Partial Pressures Radiochzmica Acta, 74 p21 Katz, J. J., Seaborg, G. T. and Morss, L. R (1986) The Chemistry of the Actinide Elements Second Edition Chapman and Hall (publishers) Kim, C.L, Cho, C.H. and Choi, H.J. (1996) Equilibnum concentration and diffision controlled leaching model for cement-based waste forms Waste Management 15 449-455 Kim, C.L, Hwang, Y., Cho, C.H. and Lee, M.C. (1993) Leach rate of radionuclides immobilized in a cylindncal waste form Proceedings of the international conference on nuclear waste managment and environmental remediation, 3, P835-840 Kim, C.L, Hwang, Y., Lee, M.C. and Kim, J. (1992) Improved leach rate experessions of radionuclides immobilized in a cylindrical cement waste form Journal of the Korean Nuclear Society 24,498-503 km,J. I., Lierse, Ch. and Baumgartner, F. (1983) Complexation of the Plutomum (IV) Ion in Carbonate - Bicarbonate Solutions in Plutonium Chemistry (ed W T Carnal1 and G R Choppin) ACS Symposium Series Volume 2 16 Kim, J.I., Bucklau, G., Rommel, H. and Sohnius (1989) The migration behaviour of transuranium elements in Gorleben Aquifer systems Colloid generation and retention process Materials Research Society Symposium, Scientific Basis for Nuclear Waste Management XI1 127, p849-854 Kim, J.I., Bukau, G. and Zhuang, W. (1987) Humic colloid generation of trcrnsuranucelements in groundwater and their migration behaviour Materials Research Society Symposium, Scientfic Basis for Nuclear Waste Management X 84, p747-756 Kim, J.I., Zeh, P., Moulin, V. and Ledoux, E. (1997) Colloid migration and modelling of colloid transport Proceedings of 4thEuropean Conference on management and disposal of radioactive waste EUR 17543 EN ~503-518 Krishnamoorthy, T.M., Joshi, S.N., Doshi, G.R. and Nair, R.N. (1992) Desorption kinetics of radionuclides fixed in cement matrix Nuclear Technology 104,35 1-357 Lagerblad, B. and Traegardh, J. (1996) Conceptual model for concrete long time degradation in a deep nuclear waste repository Swedish Cement and Concrete Research Insitute CBI report, (2 96) 105p Lasaga, A.C. (1984) Chemical kinetics of water-rock interactions Journal of Geophysical Research, 86 B6, 4009-4025 Lea, F.M (1980) The Chemistry of cement and concrete Edward Arnold (publishers) ,727~ Page 51 of 57 . - Buried Concrcte at WETS-A Literature RCVICW OBNFLInc sp- Lee, J.H., Roy, D.M., Mann, B. and Stahl, D. (1995) Integrated approach to modelling long-term durability of concrete engineered bmers in LLRW disposal facility Materials Research Soczety Symposzum, Sczentlfic Baszs for Nuclear Waste Management WIII 353, p 881-889 Lefebvre, Y.; Jolicoer, C.; Page, M.; Seabrook, P. T. (1997) Degradation Kinetics of Portland Cement Pastes, Mortars and Concretes in Acidic Environments ACI Special Publications, 1997, vol 170, vol2, p 1487-1 5 10 Letolle, R,Gegout, P., Rafai, N. and Revertegat (1992) Stable isotopes ofcarbon and oxygen for the study ofcarbonatioddecarbonation processes in concretes Cement and concrete research, 22 (2-3), 235-240 Li, F.M. and Li, Z.J. (1997) Early detection ofcorrosion of reinforcing steel in structural concrete In J Mechanzsms of Chemzcal Degradation of Cement-Based Systems Scnvener, K L , Young, F , Eds E &FN Spon, London p 159- 166 Libert, M.F., Sellier, R Jouquet, G., Trescinslu, M. Spor, H. (1993) Effects ofmicroorganisms on the long-term stability ofcement and bitumen Materzals Research Soczety Symposzum, Sczentlfic Baszs for Nuclear Waste Management WI294, p 267-272 Loida, A., Grambow, B., Geckeis, H. and Dressler, P. (1995) Processes Controlling Radionuclide Release from Spent Fuel Materials Research Soczety Symposzum, Scientzfic Baszs for Nuclear Waste Management WIII 353p577 Longworth, G. Wilhns, M.A. and Ivanovich, M. (1989) Charactenzation ofnatural colloids in a shallow glecial aquifer and their radionuclide loading Materials Research Soczety Symposzum, Sczentflc Baszs for Nuclear Waste Management XI. 127, p755-762 Lunden, I. and Andersson, K. (1991) Interaction of concrete and synthetic granitic groundwater in air and in mtrogen atmosphere Radzochzmica Acta 52/53 1 7-2 1 Majumdar, A.J., Rayment, D.L. and Pettifer, K. (1988) Microstructure and microanalysis of some ancient building matenals U K Department of Environment Report, DOE/RW/89 056 McDaniel, E.W., Spence, RD. and Tallent, O.K. (1991) Research needs in cement-based waste forms Materials Research Soczety Symposzum, SczentrJic Baszs for Nuclear Waste Management XIV 212, p449-456 Meinrath, G. and Kim, J. I., (1991) The Carbonate Complexation of the Am(II1) Ion Radzochzmica Acta 52/53 p29 Meinrath, G., Klenze, R. and Kim, J. I. (1996) Direct Spectroscopic Speciation of Uranium(V1) in Carbonate Solutions Radzochzmica Acta, 74 p8 1 Miekeley, N., Jesus. H.C., Silveira, C.L.P. and Kuechler, I.L. (1989) Colloid investigations in the Pocos de Caldas natural analogue project Materials Research Soczety Symposium, Scientific Basis for Nuclear Waste Management XII 127, p83 1-842 Miyamoto, K ,Takeda, T., Muraoka, S., Wadachi, Y., and Maeda, S. (1993) Three year expenmental study of leaching Phenomena from low level radioactive homogeneous cement-based waste forms Journal of Nuclear Sczence and Technology 30 p680-691 Moroni, L. P. and Glasser, F. P. (1995) Reactions between Cement Components and U(V1) Oxide Waste Management, 15 p243 Page 52 of 57 ~BNFL Bund Concrete at WETS-A Literature Review Inc Neall, F.M., (1996) Modellmg the long term chemical evolution of cement groundwater interactions Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management XX 412, p 483 - 490 Nishi T., Matsuda, M., Chino, K. and Kikuchi, M. (1991) Effect of porous surface-layer on leaching rate of radionuclide from cementiQous waste form Journal of Nuclear Science and Technology 28 (932-940) Nitsche, H. (1991) Solubility Studies of Transuranium Elements for Nuclear Waste Disposal Pnnciples and Overview Radiochimica Acta 52/53 p3 Nitsche, H. (1992) The Importance of Transuranium Solids in Solubility Studies for Nuclear Waste Repositories Materials Research Society Symposium, Scientrfic Basis for Nuclear Waste Management XV 257 ~298-298 Ochs, M., Hager, D., Helfer, S. and Lothenbach, B. (1998) Solubility of radionuclides in fresh and leached cementitous systems at 22OC and 5OoC Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management AXI 506, p773-780 Ollivier, J.P. and Massat, M. (1992) Permeability and microstructure of concrete a review of modelling Cement and concrete research, 22 (2-3) p503-5 14 Olofsson, U., Bengtsson, M. and Allard, B. (1986) Formation and transport of americium pseudocolloids in aqueous systems Materials Research Society Symposium, Scientific Basis for Nuclear Waste Management Ix 50, ~729-736 Payne, T.E., Edis, R. and Seo, T. (1992) Radionuclide transport by groundwater colloids at the Koongarra uranium deposit Materials Research Society Symposium, Scientific Basis for Nuclear Waste Management XV 257, p 481- -487 Peric, A.D., Plecas, I. B. and Glodic, D. (1993) Effects of relative surface area and leachant composition on the 137Cs leach rate from cement waste forms Materials Research Society Symposium, ScientrJic Basis for Nuclear Waste Management WI 294, p 24 1-246 Peric, A.D., Plecas, I.B., and Pavlovic, RS. (1995) Influence of aggressive liquid environment on the physicochemical and mechmcal charactenstic of mortar-radwaste mixture form Materials Research Society Symposium, Scientijk Basis for Nuclear Waste Management XVIII 353, p907-9 12 Peric, A.D., Plecas, I.B., Pavlovic, R.S., and Pavlovic, S.D (1994) Mortar radwaste mixture form charactenzation due to Cs leach rate using accelerated ageing processes Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management WII 333, p 377 - 382 Pfingsten, W, Shiotsulu, M. (1998) Modelling a cement degradtion experiment by a hydraulic transport and chemical equilbrium coupled code Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management x;YI 506, p 805-8 12 Plecas, I.B., Peric, A.D., Drljaca, J.D., and Kostadinovic, A.M. (1992a) Mathematical modeling of physico- chemical charactenstics of cement-waste composites in radioactive waste management Journal of Radioanalytical and Nuclear Chemistry, Articles 157,95- 104 Plecas, I, Peric, A., Glodic, S., and Kostadinovic, A.M. (1992b) Comparative interpretattion of leach rate of I3'Cs and 6oCsfrom cement matrix Journal of Radioanalytical and Nuclear Chemistry, Letters 165 377-384 Plummer, L.N. and Busenberg, E (1987) Thermodynamics of aragonite-strontianite solid solutions Results from stoichiometnc solubility at 25 and 76°C Geochimica et Cosmochimica Acta 51 1393-141 1 Page 53 of 51 @BNFL . - Buried Concrete at WETS-A Literature Review Inc Pommersheim, J.M. and Cifton, J.R (1991) Models of transport processes on concrete Nuclear Regulatory Commission NUREG/CR-4269 NISTIR-4405,lO 1p Pommersheim, J.M. and Clifton, J.R, (1994) Expansion of cementitious matenals exposed to sulphate solutions Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management XVII 333, p 363 - 368 Quillin, K.C., Duerden,S.L. and Majumdar, A.J, (1994) Accelerated ageing of blended OPC cements Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management WII 333, p 34 1 - 348 Rai, D., Felmy, A. R. and Ryan, J. L (1990) Uranium (IV) Hydrolysis Constants and the Solubility of UO, xH,O (am) Inorganic Chemistry 29, p260 Ramsay, J.D.F., Avery, RG. and Russell, P.J. (1988) Physical charactenstics and sorption behaviour of colloids generated from cementitious systems Radiochimica Acta 44/45 1 19-124 Reed, S. Lolcama, J.L. and Brandstetter, E.R. (1994) An integrated degradation and structural model for predicting the service life of buried reinforced concrete structures for low- and intermediate-level radioactive waste disposal USDOE confernce Waste Management ‘94 Report WSRC-MS-93-482 COW-940225-45, 25P RMRS (1998) Statement of work for investigating concrete containing matenal left at the Rocky Flats Environmental Technology Site after Decommissioning RM807059VP- VLP-00 14-98 Roberts, A. (1998) Personal Communication “Notes from project kick-off meeting with RMRS 20th April, 1998” BNFL R&T file reference EN0072/2/5 Rogers, R.D., Hamilton, M.A. and McConnell, J.W.Jr. (1993a) Microbial-influenced cement degradtion Literature review Nuclear Regulatory Commission NUREGKR-5987 EGG-Z695,34p Rogers, R.D., Hamilton, M.A., McConnell, J.W. (1993b) The possibility for microbially influenced degradation of cement solidified low-level radioactive waste forms Materials Research Society Symposium, Scientifc Basis for Nuclear Waste Management WI294, p 26 1-266 Rogers, RD., Hamilton, M.A., Nelson, L.O., Benson J., and Green M. (1997) Evaluation of microbially influenced degradation as a method for the decontamination of radioactively contiiminated concrete Materials Research Society Symposium, Scientifc Basis for Nuclear Waste Management XX 465, p3 17 - 324 Rogers, R.D., Hamilton, M.A., Veeh, RH and McConnell, J.W., (1994) Development of methodology to evaluate microbially influenced degradation of cement solidified low level radioactive waste forms Materials Research Society Symposium, Scientlfic Basis for Nuclear Waste Management XVII 333, p 349 - 356 Rogers, R.D., Hamilton, M.A., Veeh,R.H., and McConnel, J.W., (1996) A procedure to evaluate the potential for microbially influenced degradation of cement-solidified low-level radioactive waste forms Materials Research Society Symposium, Scientlfc Basis for Nuclear Waste Management XH 412, p 475 - 482 Roy, D.M. and Jiang, W. (1997) Concrete chemical degradation Ancient analogues and modern evaluation In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L ,Young, J F ,Eds E &FN Spon, London p 14-21 Page 54 of 57 @BNFL Buried Concrete at WETS-A Literature Review Inc Sarott, F.A., Bradbury, M.H., Pandolfo, P., and Spieler, P. (1992) Diffision and Adsorption studies of hardened cement paste and the effect ofcarbonation on diffusion rates Cement and concrete research, 22 (2- 3) 439-444 Serne, R.J., Rai, D., Martin, P.F., Felmy, A.R, Rao, L., and Ueta S. (1996) Leachability ofNd, U, Th and Sr from cements in a CO, free atmosphere Materzals Research Society Symposzum, Sczentzfic Basis for Nuclear Waste Management XIX 412, p459 - 467 Seveque, J.L., de Cayeux, M.D., Elert, M. and Nuguier, H. (1992) Mathematical modelling of radioactive waste leaching Cement and concrete research, 22 (2-3) 477-488 Silva, R. J., and Nitsche, H. (1995) Actimde Environmental Chemsistry Radzochrmrca Acta 70171, p377 Silverman, M. P. and Monoz, E. F. (1970). Fungal attack on rock solubilization and altered infrared spectra Sczence, 169,985-987 Smith, D. K., (1983) Uraruum Mineralogy in Uranium Geochemistry, Mineralogy Geology, Exploration and Resources (ed G Capald, B DeVito and F Ippolito) Institute of Mimng and Metallurgy Smith, D. W. and Strohl, W. R (1991). The Sulphur Oxidising Bactena Variations zn Autotrophic Lfe, Eds J M Shiverley and L L Barton, pp 12 1 - 146, Academic Press, London, UK Smith, R.W. and Walton, J.C. (1991) The effects of calcite solid solubon formation on the transient release of radionuclides from concrete barriers Materzals Research Society Symposzum, Sczentflc Basis for Nuclear Waste Management XIV 212, p 403- -409 Snyder, K.A. and Clifton, J.R. (1995) 4SIGHT manual a computer program for modeling degradation of underground low level waste concrete vaults US Department of Commerce Report NSTIR 5612,80p Snyder, K.A., Clifton, J.R. and Pommersheim J., (1996) A computer model to facilitate performance assessment of underground low-level waste concrete vaults Materzals Research Socrety Symposzum, Scientflc Basis for Nuclear Waste Management XZX 412, p 491 - 498 Sowder, A. G., Clark, S. B. and Fjeld, R A. (1996) The Effect of Silica and Phosphate on the Transformation of Schoepite to Becquerelite and other Uranyl Phases Radzochzmzca Acta 74 Stenhouse, M. J. (1995), Sorption Databases for Crystalline, Marl and Bentomte for Performance Assessments, NAGRA Technical Report 93 - 06 Stumm, W. and Morgan, J.J. (1981) Aquatzc Chemzstry John Wiley & Sons 780p Sverjenslcy, D., Bennett, D. G. and Read, D (1992) Geochemical Modelling of Secondary Uranium Ore Formation Alligator fivers Analogue Project Final Report Volume 11 Taylor, H.F.W. and Gollop, R.S. (1997) Some chemical and microstructural aspects of concrete durability In Mechanisms of Chemical Degradation of Cement-Based Systems Scrivener, K L , Young, J F , Eds E &FN Spon, London pl77- 184 Thomassin, J.H and Rassineux, F (1992) Ancient analogues of cement-based materials stability of calcium silicate hydrates Applied Geochemistry Suppl Issue 1 1 37- 142 Torrero, M. E., Casas, I., de Pablo, J., Sandino, M. C. A. and Grambow, B. (1994) A Compmson between Unirradiated UO,(s) and Schoepite Solubilities in a 1M NaCl Medium Radzochzmzca Acta 66/67 p29 Turrrziani R. (1986) Internal degradation of concrete alkali-aggregate reaction, reinforcement steel corrosion (An), Congr Int Quim Cimento, 8"' Volume 1, 388-442 Page 55 of 57 @BNFL Buncd Concrete at WETS-A Literature Review Inc ------Tuum, K. (1982) Corrosion of steel in concrete, CBI Research 469p 4-82 Vejmelka, P., Rudolph, G. and Kluger, W. (1991) Investigation of low and medium - level cemented waste forms from reprocessing Charactensation of radioactive waste forms A senes of final reports (1 985 - 89) No 4, Commission of the European Communities Report EUR 13544 EN Viellard, P. and Rassineux, F (1992) Thermodynamic and geochemical modelling of the alteration of two cement matnces Applied Geochemistry Suppl Issue 1 125-136 Waite, T. D., Davis, J. A., Payne, T. E., Waychunas, G. A. and Xu, N. (1994) Uranium (VI) adsorption to femhydrite Application of a Surface Complexation Model Geochimica et Cosmochimica Acta 24 p5465 Walton, J.C. (1992) Peformance of intact and partially degraded concrete bmers in limiting mass transport Nuclear Regulatory Commission NUREG/CR-5445,3 9p Walton, J.C. and Otis, M.D. (1990) Concrete barrier performance in radioactive waste disposal in the unsaturated zone Materials Research Society Symposium, Scient@ Basis for Nuclear Waste Management XIII 176, ~711-717 Walton, J.C., Smith, R.W., Tarquin, A., Sheeley, P. Kalyana, R., Gwynne, J, Gutierrez, N, Bin-shafique, M.S., Rodriguez, M. and Andrade, R. (1997) Role of carbonation in long-term performance of cementitious wasteforms In Mechanisms of Chemical Degradation of Cement-Based Systems Scrivener, K L ,Young, J F , Eds E &FN Spon, London p3 15-322 Wang and Gillott (1991) Mechanism of alkali-silica reaction and the significance of calcium hydroxide Cement and Concrete Research 21,647-654 Webley, D. M., Henderson, E. K. and Taylor, F (1963) The microbiology of rocks and weathered stone, J soil SC~,14, 102-1 12 Wheat, H.G., Kasthurirangan, J. and Kitowski, C.J. (1997) Behaviour of steel in simulated concrete solutions (1 997) In Mechanisms of Chemical Degradation of Cement-Based Systems Scnvener, K L , Young, J F , Eds E &FN Spon, London p143-150 White, C., Sayer, J. A. and Gadd, G. M. (1997) Microbial solubilisation and immobilization of toxic metals key biogeochemical processes for treatment of contaminants FEMS Microbiological Reviews, 20, 503-5 16 Wieland, E. Tits, J.T., Spieler, P. and Dobler, J.P. (1998) Interaction of Eu(II1) and Th(1V) with sulphate- resisting Portland cement Materials Research Society Symposium, Scientific Basis for Nuclear Waste Management XXI 506, p573-578 Wronluewicz, D. J., Bates, J. K., Gerding, T. J., Veleckis, E. and Tani, B. S. (1992) Uranium release and Secondary Phase Formation during Unsaturated Testing of UO, at 9OoC Journal of Nuclear Materials 190 p107 Yajima, T., Kawamura, Y. and Ueta, S (1995) Uranium (IV) Solubility and Hydrolysis Constants under Reduced Conditions Materials Research Society Symposium, ScientiJic Basis for Nuclear Waste Management WIIL353 ~1137-1142 Yamaguchi, T., Sakomoto, Y. and Ohnulu, T. (1994) Effect of the Complexation on Solubility of Pu(1V) in aqueous Carbonate Systems Radiochimica Acta 66/67 p9 Yamamura, T., Kitamura, A., Fukui, A., Nishikawa, S., Yamamoto, T. and Monyama, H. (1997) Solubility of U(V1) in Highly Basic Solutions Presented at the Sixth International Conference on the Page 56 of57 Buried Concrete at WETS-A Literature Review P , Chemistry and Migration Behaviour of Actmides and Fission Products in the Geosphere (Migration '97), I Sendai, Japan, October, 1997 I Yang, X.N.; Zhou, W.Y.; Wang, RS. (1995) Study of the immobilization of 226Ra I1 Predictive model of radionuclide leaching behavior in a cement-based solid Journal of Radioanalytical and Nuclear Chemistry (Hungary) , 19812 (287-294) Page 57 of 51